RU2822503C9 - Method, equipment and device for encoding and decoding - Google Patents

Abstract

FIELD: image processing.

SUBSTANCE: invention relates to image encoding/decoding. Disclosed is a method of encoding images, comprising: when it is determined that weighted prediction should be allowed for a current block, the method includes: obtaining a weighted prediction angle of the current block for the current block; for each pixel position of the current block, determining a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block; determining the target weight value of the pixel position based on the reference weight value associated with the surrounding corresponding position; determining the associated pixel position weight based on the target pixel position weight; determining a first pixel position prediction value based on a first prediction mode of the current block; determining a second pixel position prediction value based on a second prediction mode of the current block; and based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determining the pixel position weighted prediction value; and determining weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

EFFECT: high efficiency of encoding/decoding.

22 cl, 21 dwg, 10 tbl

Description

AREA OF TECHNOLOGY

[0001] The present invention relates to methods, equipment and devices for encoding and decoding.

PREREQUISITES FOR THE CREATION OF THE INVENTION

[0002] In order to reduce the occupied volume, video or images are transmitted after encoding. The video encoding method may include processes such as prediction, transformation, quantization, entropy encoding, filtering, etc. Predictive encoding may include intra-coding and inter-coding. Inter-coding refers to the operation of using temporal correlation of video to predict current pixels of the current image using pixels of a neighboring encoded image to achieve effective elimination of temporal redundancy of video. Intra-coding refers to the operation of using spatial correlation of video to predict current pixels using pixels of one or more encoded blocks of the current image to achieve effective elimination of spatial redundancy of video.

ESSENCE OF THE INVENTION

[0003] The present invention provides methods, equipment and devices for encoding and decoding.

[0004] The present invention provides a method for encoding and decoding. When it is determined that weighted prediction should be enabled for a current block, the method includes: obtaining a weighted prediction angle of the current block for the current block; for each pixel position of the current block, determining a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block; determining a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position, determining an associated weight value of the pixel position based on the target weight value of the pixel position; for each pixel position of the current block, determining a first prediction value of the pixel position based on a first prediction mode of the current block, determining a second prediction value of the pixel position based on a second prediction mode of the current block; and based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determining a weighted prediction value of the pixel position; and determining the weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

[0005] The present invention provides an encoding device comprising a module configured to obtain, for a current block, when it is determined that weighted prediction should be enabled for the current block, a weighted prediction angle of the current block; a module configured to determine, for each pixel position of the current block, a surrounding corresponding position indicated by the pixel position, from surrounding positions outside the current block, based on the weighted prediction angle of the current block; a module configured to determine a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position, and to determine an associated weight value of the pixel position based on the target weight value of the pixel position; the module configured to determine, for each pixel position of the current block, a first prediction value of the pixel position based on a first prediction mode of the current block, to determine a second prediction value of the pixel position based on a second prediction mode of the current block; and, based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determine a weighted prediction value of the pixel position; and a module configured to determine weighted prediction values of the current block based on weighted prediction values of all pixel positions of the current block.

[0006] The present invention provides a decoding device comprising a module configured to obtain, for a current block, when it is determined that weighted prediction should be enabled for the current block, a weighted prediction angle of the current block; a module configured to determine, for each pixel position of the current block, a surrounding corresponding position indicated by the pixel position, from surrounding positions outside the current block, based on the weighted prediction angle of the current block; a module configured to determine a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position, and to determine an associated weight value of the pixel position based on the target weight value of the pixel position; the module configured to determine, for each pixel position of the current block, a first prediction value of the pixel position based on a first prediction mode of the current block, to determine a second prediction value of the pixel position based on a second prediction mode of the current block; and, based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determine a weighted prediction value of the pixel position; and a module configured to determine weighted prediction values of the current block based on weighted prediction values of all pixel positions of the current block.

[0007] The present invention provides a decoder-side device comprising a processor and a machine-readable storage medium on which machine-readable instructions executable by the processor are stored. The processor is configured to execute machine-readable instructions for performing the following: when it is determined that weighted prediction should be enabled for a current block, obtaining for the current block a weighted prediction angle of the current block; for each pixel position of the current block, determining a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block; determining a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position, determining an associated weight value of the pixel position based on the target weight value of the pixel position; for each pixel position of the current block, determining a first prediction value of the pixel position based on a first prediction mode of the current block, determining a second prediction value of the pixel position based on a second prediction mode of the current block; and based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determining a weighted prediction value of a pixel position; and determining weighted prediction values of a current block based on weighted prediction values of all pixel positions of the current block.

[0008] The present invention provides an encoder-side device comprising a processor and a machine-readable storage medium on which machine-readable instructions executable by the processor are stored. The processor is configured to execute the machine-readable instructions for performing the following: when it is determined that weighted prediction should be enabled for a current block, obtaining for the current block a weighted prediction angle of the current block; for each pixel position of the current block, determining a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block; determining a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position, determining an associated weight value of the pixel position based on the target weight value of the pixel position; for each pixel position of the current block, determining a first prediction value of the pixel position based on a first prediction mode of the current block, determining a second prediction value of the pixel position based on the second prediction mode of the current block; and based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determining a weighted prediction value of a pixel position; and determining weighted prediction values of a current block based on weighted prediction values of all pixel positions of the current block.

[0009] As described above, in the embodiments of the present invention, an efficient method for setting weight values is provided in which a suitable target weight value is set for each pixel position of a current block, so that the prediction values of the current block are closer to the original pixels, which improves the prediction accuracy, the prediction efficiency, and the coding efficiency.

BRIEF DESCRIPTION OF DRAWINGS

[00010] Fig. 1 is a diagram of the structure of a video coding system.

[00011] Figs. 2A-2C are schematic diagrams of weighted prediction.

[00012] Fig. 3 is a block diagram illustrating a method for encoding and decoding according to an embodiment of the present invention.

[00013] Fig. 4A is a block diagram illustrating a coding method according to an embodiment of the present invention.

[00014] Figs. 4B-4E are diagrams illustrating surrounding positions beyond the current block.

[00015] Fig. 4F is a block diagram illustrating a decoding method according to an embodiment of the present invention.

[00016] Fig. 5A and 5B are diagrams illustrating a weighted prediction angle according to an embodiment of the present invention.

[00017] Fig. 6A is a block diagram of determining a reference weight value of an ambient position according to an embodiment of the present invention.

[00018] Fig. 6B is a block diagram of obtaining a weight array according to an embodiment of the present invention.

[00019] Figs. 6C-6D are schematic diagrams of the corners of the current block in various regions according to an embodiment of the present invention.

[00020] Fig. 7 is a diagram illustrating neighboring blocks of a current block according to an embodiment of the present invention.

[00021] Fig. 8A is a structural schematic diagram illustrating an encoding and decoding device according to an embodiment of the present invention.

[00022] Fig. 8B is a diagram of a hardware structure of a decoder-side device according to an embodiment of the present invention.

[00023] Fig. 8C is a diagram of a hardware structure of an encoder-side device according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[00024] The terms used herein are intended to describe a particular embodiment of the invention only and are not intended to limit the present invention. The singular forms used in the present specification and the appended claims are also intended to include the plural forms unless the context clearly dictates otherwise. It should also be understood that the phrase "and/or" as used herein refers to any and all possible combinations that include one or more of the related listed elements. It should be noted that while the terms "first," "second," "third," and the like may be used in the present invention to describe various information, such information should not be limited to these terms. These terms are used merely to distinguish one category of information from another. For example, within the spirit of the present invention, first information may be referred to as second information; and, similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may mean "when," or "after," or "in response to a determination."

[00025] The present invention provides methods, equipment and devices for encoding and decoding, which may include the following concepts: intra prediction, inter prediction and intra block copy (IBC) prediction.

[00026] Intra prediction uses spatial correlation of video to predict the current pixel using pixels of one or more coded blocks of the current image to eliminate spatial redundancy of video. Intra prediction defines several prediction modes, each of which corresponds to one texture direction (except for the direct current (DC) mode). For example, if the image texture corresponds to the horizontal direction, using the horizontal prediction mode can better predict image information.

[00027] Inter prediction uses pixels of neighboring coded images to predict pixels of the current image based on temporal correlation of the video due to strong temporal correlation included in the video sequence in order to effectively eliminate temporal redundancy of the video. Inter prediction in video coding standards uses block-based motion compensation technology, in which the best corresponding block is found for each block of pixels of the current image in one or more previously coded images. This process is called Motion Estimation (ME).

[00028] Intra-block copying allows referencing the same slice if the reference data of the current block comes from the same slice. In the intra-block copying technology, a prediction value of the current block can be obtained using a block vector of the current block. For example, based on the characteristic that there are a large number of repeated textures in the same slice in the screen content, when the prediction value of the current block is obtained using the block vector, the compression efficiency of the screen content sequence can be improved.

[00029] The prediction pixel refers to a pixel value obtained from an encoded/decoded pixel, and a residual is obtained based on the difference between the original pixel and the prediction pixel, and then transformation, quantization, and coding of the residual coefficients are performed. The inter prediction pixel for the current block refers to a pixel value obtained from a reference image, and since the pixel positions are discrete, an interpolation operation is required to obtain the final prediction pixel. The closer the prediction pixel is to the original pixel, the smaller the residual energy obtained by performing subtraction for both pixels, and the higher the compression efficiency of coding.

[00030] Motion Vector (MB). In inter coding, a motion vector is used to represent the relative displacement between the current block of the current picture and the reference block of the reference picture. Each of the divided blocks has a corresponding motion vector, which must be transmitted to the decoder side. If the motion vector of each block is coded and transmitted independently, especially when there are a large number of smaller blocks, more bits must be used. In order to reduce the number of bits used to code the motion vector, spatial correlation between neighboring blocks can be used to predict the motion vector of the current block to be coded based on the motion vector(s) of the neighboring coded block(s), and then the prediction difference is coded. In this way, the number of bits representing motion vectors can be effectively reduced. Based on this, in the process of encoding the motion vector of the current block, first one or more motion vectors of one or more neighboring coded blocks are used to predict the motion vector of the current block, and then the motion vector difference (MVD) between the motion vector prediction value (MVP) and the actual motion vector estimate value is encoded.

[00031] Motion information. Since a motion vector is a position offset from a current block to a reference block, in order to accurately obtain information about a specified block, in addition to the motion vector, index information about a reference picture is also needed to indicate which reference picture is used for the current block. In a video coding technology, a reference picture list may be set for the current picture, and the index information about the reference picture indicates which reference picture specified in the reference picture list is used for the current block. In addition, many coding technologies also support multiple reference picture lists, so that an index may also be used to indicate which reference picture list is used, and this index may be referred to as a reference direction. To summarize, in a video coding technology, motion-related information such as a motion vector, index information of a reference picture, and a reference direction may be collectively referred to as motion information.

[00032] Block vector (BV). Block vector is applied in intra-block copy technology that uses block vector for motion compensation, for example, the prediction value of the current block is obtained using block vector. Block vector represents the relative offset between the current block and the best corresponding block from the coded blocks of the current slice, and is different from the motion vector. Based on the characteristic of having a large number of repeated textures in the same slice, when the prediction value of the current block is obtained using block vector, the compression efficiency can be improved.

[00033] Intra prediction mode. In intra coding, the intra prediction mode is used to perform motion compensation, for example, the prediction value of the current block is obtained using the intra prediction mode. For example, the intra prediction mode may include, but is not limited to, a planar mode, a DC mode, and 33 corner modes. Table 1 shows examples of the intra prediction mode, where the planar mode corresponds to mode 0, the DC mode corresponds to mode 1, and the remaining 33 corner modes correspond to modes 2 through 34. The planar mode is applicable to a region in which pixel values change slowly, and uses two linear filters in the horizontal and vertical directions to obtain an average pixel value in two directions as a prediction value of the pixel of the current block. The DC mode is applicable to a large flat surface and takes an average value of the surrounding pixels of the current block as a prediction value of the pixel of the current block. 33 corner modes are available. The next generation VVC (Versatile Video Coding) encoding and decoding standard uses split-angle modes, such as 65-angle modes.

[00034] Palette mode. In the palette mode, the pixel values of the current block are represented by a small set of pixel values, i.e., a palette. When the pixel value of a pixel position in the current block is close to a color in the palette, the pixel position is encoded by the index value of the corresponding color in the palette. When the pixel value of a pixel position in the current block is not similar to all colors in the palette, the pixel position must be encoded by an escape pixel, and the escape pixel is directly quantized and then encoded into the bitstream. At the decoder side, one palette is first obtained, for example, a palette storing {color A, color B, color C}, and then it is determined whether each pixel position is encoded as an escape pixel. If the pixel position is not a runaway pixel, the pixel position index is obtained from the bitstream, and then a color based on the pixel position index is obtained from the palette and assigned to the pixel position; otherwise, the runaway pixel is analyzed.

[00035] Rate-Distortion Optimization (RDO). There are two metrics for evaluating coding efficiency: bit rate and Peak Signal to Noise Ratio (PSNR). The smaller the bit rate, the higher the compression ratio; and the higher the PSNR, the better the quality of the reconstructed image. When selecting a mode, the decision method is actually a comprehensive evaluation for both modes. For example, the cost corresponding to a mode can be calculated by the following formula: J/(mode)=D+λ*R, where D represents the distortion, usually measured by the sum of squared errors (SSE), and SSE refers to the average sum of squared differences between the reconstructed image block and the original image; λ represents the Lagrange multiplier; and R represents the actual number of bits required to encode an image block in that mode, including the total number of bits required for the encoding mode information, motion information, residuals, etc. When selecting a mode, if RDO is used to make a comparative decision about encoding modes, the best encoding efficiency can usually be guaranteed.

[00036] Structure of a video coding system. The structure of the video coding system shown in Fig. 1 can be used to perform a process on the encoder side in the embodiments of the present invention, and the structure of the video decoding system is similar to Fig. 1 and will not be repeated here. The structure of the video decoding system can be used to perform a process on the decoder side in the embodiments of the present invention. For example, the structure of the video coding system and the structure of the video decoding system may include, but are not limited to, an intra prediction unit, a motion estimation/compensation unit, a reference picture buffer, an in-loop filtering unit, a reconstruction unit, a transform unit, a quantization unit, an inverse transform unit, a dequantization unit, an entropy coder, and the like. The encoder can perform the process on the encoder side through the interaction of these units, and the decoder can perform the process on the decoder side through the interaction of these units.

[00037] For example, the current block may be rectangular, while the boundary of an object is usually not rectangular in practical situations. Therefore, two different entities are usually used for the boundary of an object (for example, an object represented in the foreground and in the background, etc.). When the movements of two objects are incompatible, the two objects cannot be well separated based on the rectangular division. Therefore, the current block may be divided into two non-rectangular sub-blocks, and weighted prediction is performed on the two non-rectangular sub-blocks. For example, weighted prediction is intended to perform a weighted operation based on multiple prediction values to obtain one or more final prediction values. Weighted prediction may include: combined inter/intra prediction, combined inter/intra prediction, combined intra/intra prediction, etc. For the weight(s) of a weighted prediction, the same weight value or different weight values may be set for all pixel positions of the current block.

[00038] Fig. 2A is a schematic diagram illustrating a Triangular Partition Mode (TPM) of inter weighted prediction.

[00039] The TPM prediction block is obtained by combining the inter prediction block 1 (for example, the inter prediction value 1 for multiple pixel positions is obtained using the inter prediction mode) and the inter prediction block 2 (for example, the inter prediction value 2 of multiple pixel positions is obtained using the inter prediction mode). The TPM prediction block may be divided into two regions/parts, one of which may be the inter prediction region 1, and the other may be the inter prediction region 2. The TPM prediction block may be rectangular, and the two inter prediction regions of the TPM prediction block may be non-rectangular, and the boundary line between the two inter prediction regions (as shown by the dotted line in Fig. 2B) may be the main diagonal or sub-diagonal of the TPM prediction block.

[00040] For example, the position of each pixel in the inter prediction region 1 is mainly determined based on the inter prediction value of the inter prediction block 1. For example, the inter prediction value of the inter prediction block 1 of the pixel position and the inter prediction value of the inter prediction block 2 of the pixel position are weighted to obtain a combined pixel position prediction value, where the inter prediction value of the inter prediction block 1 has a larger weight value, and the inter prediction value of the inter prediction block 2 has a smaller weight value or even 0. Each pixel position of the inter prediction region 2 is mainly determined based on the inter prediction value of the inter prediction block 2. For example, the inter prediction value of inter prediction block 1 of the pixel position and the inter prediction value of inter prediction block 2 of the pixel position are weighted to obtain a combined pixel position prediction value, wherein the inter prediction value of inter prediction block 2 has a larger weight value, and the inter prediction value of inter prediction block 1 has a smaller weight value or even 0. Finally, the combined pixel position prediction values are combined to form a TPM prediction block.

[00041] Fig. 2B is a schematic diagram illustrating a Geometrical (GEO) division mode for an inter-block mode. The GEO mode is used to divide an inter prediction block into two sub-blocks using one division line. In the GEO mode, there may be more division directions compared to the TPM mode. The GEO mode is similar to the TPM mode in the weighted prediction process.

[00042] The GEO prediction block is obtained by combining the inter prediction block 1 (for example, the inter prediction value 1 for the positions of several pixels is obtained using the inter prediction mode) and the inter prediction block 2 (for example, the inter prediction value 2 for the positions of several pixels is obtained using the inter prediction mode). The GEO prediction block may be divided into two regions/parts, one of which may be the inter prediction region 1 and the other may be the inter prediction region 2.

[00043] For example, the position of each pixel in the inter prediction region 1 is mainly determined based on the inter prediction value of the inter prediction block 1. For example, the inter prediction value of the inter prediction block 1 of the pixel position and the inter prediction value of the inter prediction block 2 of the pixel position are weighted to obtain a combined pixel position prediction value, wherein the inter prediction value of the inter prediction block 1 has a larger weight value, and the inter prediction value of the inter prediction block 2 has a smaller weight value or even 0. Each pixel position of the inter prediction region 2 is mainly determined based on the inter prediction value of the inter prediction block 2. For example, the inter prediction value of inter prediction block 1 of the pixel position and the inter prediction value of inter prediction block 2 of the pixel position are weighted to obtain a combined pixel position prediction value, wherein the inter prediction value of inter prediction block 2 has a larger weight value, and the inter prediction value of inter prediction block 1 has a smaller weight value or even 0. Finally, the combined pixel position prediction values are combined to form a GEO prediction block.

[00044] For example, weight values of the GEO prediction block are set taking into account the distance of the pixel position from the dividing line. As shown in Fig. 2C, the pixel position A, the pixel position B, and the pixel position C are located at the lower right of the dividing line, and the pixel position D, the pixel position E, and the pixel position F are located at the upper left of the dividing line. For the pixel position A, the pixel position B, and the pixel position C, the weight values of the inter prediction block 2 are ranked as B≥A≥C, and the weight values of the inter prediction block 1 are ranked as C≥A≥B. For the pixel position D, the pixel position E, and the pixel position F, the weight values of the inter prediction block 1 are ranked as D≥F≥E, and the weight values of the inter prediction block 2 are ranked as E≥F≥D. The above method requires calculating the distance between the pixel position and the dividing line, and then determining the weight value of the pixel position.

[00045] In the above examples, in order to achieve weighted prediction, it is necessary to determine one or more weight values of a prediction block corresponding to each pixel position of the current block, and weighted prediction is performed based on one or more weight values corresponding to each pixel position. But the setting of one or more weight values depends on the dividing line, and unreasonable setting of one or more weight values may result in poor prediction results and low coding efficiency.

[00046] In view of this, embodiments of the present invention propose a method for obtaining one or more weight values. In this way, based on reference weight values of surrounding positions outside the current block, a target weight value is determined for each pixel position of the current block, so that a more suitable target weight value can be set for each pixel position. In this case, the prediction values of the current block are closer to the original pixels, which improves the prediction accuracy, the prediction efficiency, and the coding efficiency.

[00047] The encoding and decoding method according to embodiments of the present invention will be described in detail below in conjunction with several specific embodiments of the invention.

[00048] Embodiment 1 of the invention. Fig. 3 is a flow chart illustrating a method of encoding and decoding according to embodiments of the present invention. The method of encoding and decoding is applied on the decoder side or on the encoder side. The method of encoding and decoding includes the following steps 301-304.

[00049] In step 301, when it is determined that weighted prediction should be enabled for the current block, a weighted prediction angle of the current block is obtained for the current block.

[00050] In some examples, the current block may include one sub-block, i.e., the sub-block is the current block itself. In this case, obtaining the weighted prediction angle of the sub-block means obtaining the weighted prediction angle of the current block.

[00051] For example, when predicting the current block, the decoder side or the encoder side first determines whether to enable weighted prediction for the current block. If it is determined that weighted prediction should be enabled for the current block, the weighted prediction angle of the current block is obtained using the encoding and decoding method according to the embodiments of the present invention. If it is determined that weighted prediction should not be enabled for the current block, no limitations are made on the corresponding implementations in the embodiments of the present invention.

[00052] For example, when it is determined that weighted prediction should be enabled for the current block, a weighted prediction angle of the current block should be obtained, wherein the weighted prediction angle refers to an angular direction indicated by a pixel position within the current block. For example, the angular direction indicated by a pixel position within the current block is determined based on a determined weighted prediction angle, wherein the angular direction points to a surrounding position outside the current block.

[00053] At step 302, for each pixel position of the current block, a surrounding corresponding position indicated by the pixel position is determined from surrounding positions outside the current block based on a weighted prediction angle of the current block; a target weight value of the pixel position is determined based on a reference weight value associated with the surrounding corresponding position; and an associated weight value of the pixel position is determined based on the target weight value of the pixel position.

[00054] For example, since a weighted prediction angle refers to an angular direction indicated by a pixel position within a current block, for each pixel position of the current block, the angular direction indicated by the pixel position is determined based on the weighted prediction angle, and then a surrounding corresponding position indicated by the pixel position is determined from surrounding positions outside the current block based on the angular direction.

[00055] For each pixel position of the current block, after the surrounding corresponding position indicated by the pixel position is determined, a reference weight value associated with the surrounding corresponding position is determined, wherein the reference weight value associated with the surrounding corresponding position may be pre-configured or determined based on a strategy that is not limited to the present invention, provided that the surrounding corresponding position has an associated reference weight value.

[00056] Then, based on the reference weight value associated with the surrounding corresponding position, a target weight value of the pixel position is determined, for example, the reference weight value associated with the surrounding corresponding position can be determined as the target weight value of the pixel.

[00057] After obtaining the target pixel position weight value, the associated pixel position weight value may be determined based on the target pixel position weight value. For example, the sum of the target weight value and the associated weight value for each pixel position may be a fixed predetermined value. Therefore, the associated weight value may be the difference between the predetermined value and the target weight value. Assuming that the predetermined value is 8 and the target pixel position weight value is 0, the associated pixel position weight value is 8; assuming that the target pixel position weight value is 1, the associated pixel position weight value is 7, and so on, provided that the sum of the target weight value and the associated weight value is 8.

[00058] At step 303, for each pixel position of the current block, a first prediction value of the pixel position is determined based on a first prediction mode of the current block, a second prediction value of the pixel position is determined based on a second prediction mode of the current block; and a weighted prediction value of the pixel position is determined based on the first prediction value, the target weight value, the second prediction value, and the associated weight value of the pixel position.

[00059] For example, if the target weight value is a weight value corresponding to a first prediction mode, and the associated weight value is a weight value corresponding to a second prediction mode, the weighted pixel position prediction value may be as follows: (first pixel position prediction value * target pixel position weight value + second pixel position prediction value * associated pixel position weight value) / a fixed predetermined value.

[00060] In addition, if the target weight value is a weight value corresponding to the second prediction mode, and the associated weight value is a weight value corresponding to the first prediction mode, the weighted pixel position prediction value may be as follows: (second pixel position prediction value * target pixel position weight value + first pixel position prediction value * associated pixel position weight value) / a fixed predetermined value.

[00061] At step 304, weighted prediction values of the current block are determined based on weighted prediction values of all pixel positions of the current block.

[00062] For example, for the current block, the weighted prediction values of all pixel positions of the current block are formed into the weighted prediction values of the current block.

[00063] As described above, in the embodiments of the present invention, an efficient method for setting weight values is provided, in which a suitable target weight value is set for each pixel position of a current block, so that the prediction values of the current block are closer to the original pixels, which improves the prediction accuracy, the prediction efficiency, and the coding efficiency.

[00064] Embodiment 2 of the invention. In embodiment 1 of the invention, the current block may include one sub-block, i.e., the sub-block is the current block itself. For the current block, Fig. 4A shows a schematic diagram illustrating a block diagram of the encoding method. The method is applied on the encoder side and includes the following steps 401-407.

[00065] In step 401, when it is determined that weighted prediction should be enabled for the current block, the encoder side obtains a weighted prediction angle and a weighted prediction position of the current block.

[00066] For example, on the encoder side, it is necessary to determine whether to enable weighted prediction for the current block. If the encoder side determines that weighted prediction should be enabled for the current block, the weighted prediction angle and the weighted prediction position of the current block are obtained and subsequent steps are performed, otherwise the processing method is not limited in the present invention.

[00067] In a possible implementation, if the current block satisfies the condition for enabling weighted prediction, the encoder side may determine that weighted prediction should be enabled for the current block. If the current block does not satisfy the condition for enabling weighted prediction, the encoder side may determine that weighted prediction should not be enabled for the current block. For example, the encoder side may determine whether the characteristic information of the current block satisfies a certain condition. If the characteristic information of the current block satisfies a certain condition, it is determined that weighted prediction should be enabled for the current block, and if not, it is determined that weighted prediction should not be enabled for the current block.

[00068] For example, the characteristic information includes, but is not limited to, one or more of the following: a type of the current slice in which the current block is located, information about the size of the current block, and switching control information. The switching control information may include, but is not limited to: sequence level switching control information (e.g., Sequence Parameter Set (SPS) and Sequence Header (SH)), picture level switching control information (e.g., Picture Parameter Set (PPS) and Picture Header (PH)), slice level switching control information (e.g., slice, tile, and patch), or largest coding unit level switching control information (e.g., Largest Coding Unit (LCU) and Coding Tree Unit (CTU)).

[00069] For example, if the characteristic information is a type of the current slice in which the current block is located, the type of the current slice in which the current block is located that satisfies a certain condition includes, but is not limited to, the following: if the type of the current slice in which the current block is located is a B slice, determining that the slice type satisfies a certain condition; or, if the type of the current slice in which the current block is located is an I slice, determining that the slice type satisfies a certain condition.

[00070] For example, if the characteristic information is information about the size of the current block, such as the width and height of the current block, the information about the size of the current block satisfying a certain condition includes, but is not limited to, the following: if the width is greater than or equal to a first value, and the height is greater than or equal to a second value, it is determined that the information about the size of the current block satisfies a certain condition; or, if the width is greater than or equal to a third value, the height is greater than or equal to a fourth value, the width is less than or equal to a fifth value, and the height is less than or equal to a sixth value, it is determined that the information about the size of the current block satisfies a certain condition; or the product of the width and the height is greater than or equal to a seventh value, it is determined that the information about the size of the current block satisfies a certain condition. The above-mentioned values can be configured based on experience, for example, configured as 8, 16, 32, 64 or 128, etc., which is not limited here. For example, the first value may be 8, the second value may be 8, the third value may be 8, the fourth value may be 8, the fifth value may be 64, the sixth value may be 64, and the seventh value may be 64. In this case, if the width is greater than or equal to 8, and the height is greater than or equal to 8, it is determined that the size information of the current block satisfies a certain condition; or, if the width is greater than or equal to 8, the height is greater than or equal to 8, the width is less than or equal to 64, and the height is less than or equal to 64, it is determined that the size information of the current block satisfies a certain condition; or, if the product of the width and the height is greater than or equal to 64, it is determined that the size information of the current block satisfies a certain condition. The above examples are merely illustrative and do not limit the present invention.

[00071] For example, if the characteristic information is information about the size of the current block, such as the width and height of the current block, the information about the size of the current block that satisfies a certain condition includes, but is not limited to, the following: the width is not less than a and not greater than b, the height is not less than a and not greater than b, where a may be less than or equal to 16, and b may be greater than or equal to 16. For example, a is 8, and b is 64 or 32.

[00072] For example, if the characteristic information is switching control information, the switching control information satisfying a certain condition includes, but is not limited to, the following: if the switching control information allows a weighted prediction to be resolved for the current block, it is determined that the switching control information satisfies the certain condition.

[00073] For example, if the characteristic information is the type of the current slice in which the current block is located and the size information of the current block, when the slice type satisfies a certain condition and the size information satisfies a certain condition, it is determined that the characteristic information of the current block satisfies the certain condition. If the characteristic information is the type of the current slice in which the current block is located and the switching control information, when the slice type satisfies a certain condition and the switching control information satisfies a certain condition, it is determined that the characteristic information of the current block satisfies the certain condition. If the characteristic information is the size information of the current block and the switching control information, when the size information satisfies a certain condition and the switching control information satisfies a certain condition, it is determined that the characteristic information of the current block satisfies the certain condition. If the characteristic information is a type of the current slice in which the current block is located, information about the size of the current block, and switching control information, when the slice type, the size information, and the switching control information satisfy certain conditions, respectively, it is determined that the characteristic information of the current block satisfies the certain condition.

[00074] In a possible implementation, when it is determined that weighted prediction should be enabled for the current block, the encoder side obtains a weighted prediction angle and a weighted prediction position of the current block. In Fig. 4B, based on the weighted prediction angle, an angular direction indicated by a pixel position (e.g., pixel position 1, pixel position 2, and pixel position 3) within the current block is shown, wherein the angular direction points to a surrounding position outside the current block. In Fig. 4C, based on another weighted prediction angle, an angular direction indicated by a pixel position (e.g., pixel position 2, pixel position 3, and pixel position 4) within the current block is shown, wherein the angular direction points to a surrounding position outside the current block.

[00075] A weighted prediction position (also called a distance parameter) is used to configure a reference weight value of a surrounding position outside the current block. For example, based on parameters such as a weighted prediction angle, a size of the current block, etc., a range of surrounding positions outside the current block is determined, as shown in Figs. 4B and 4C. The range of surrounding positions is equally divided into N parts, where the value of N can be set arbitrarily, for example, can be set to 4, 6, or 8, etc. The weighted prediction position is used to represent which surrounding position is used as the starting position of the weight transformation of the current block, so that the reference weight values of the surrounding positions outside the current block are configured based on the starting position of the weight transformation.

[00076] For example, the description is given for N=8. As shown in Fig. 4D, after all the surrounding positions are divided into eight equal parts, seven weighted prediction positions can be obtained. When the weighted prediction position is 0, it indicates that the surrounding position ao (that is, the surrounding position marked with the dotted line 0. In practice, there is no dotted line 0, and the dotted line 0 is only used to facilitate the understanding of these examples. The dotted lines from 0 to 6 are used to equally divide all the surrounding positions into eight parts) serves as the starting position of the weight transformation of the current block. When the weighted prediction position is 1, it indicates that the surrounding position a ₁ serves as the starting position of the weight transformation of the current block. Similarly, when the weighted prediction position is 6, it means that the surrounding position ae is used as the starting position of the weight transformation of the current block.

[00077] For example, for different weighted prediction angles, the values of N may be different. For example, for weighted prediction angle A, the value of N may be 6, which indicates that the range of surrounding positions determined based on weighted prediction angle A is equally divided into six parts. For weighted prediction angle B, the value of N may be 8, which indicates that the range of surrounding positions determined based on weighted prediction angle B is equally divided into eight parts.

[00078] For different weighted prediction angles, the values of N may be the same. In the case of the same value of N, different numbers of weighted prediction positions may be selected. For example, for weighted prediction angle A, the value of N is 8, which indicates that the range of surrounding positions determined based on weighted prediction angle A is equally divided into eight parts; and for weighted prediction angle B, the value of N is 8, which indicates that the range of surrounding positions determined based on weighted prediction angle B is equally divided into eight parts. At the same time, the weighted prediction positions corresponding to weighted prediction angle A may be selected from a total of five positions from a ₁ to a ₅ , and the weighted prediction positions corresponding to weighted prediction angle B may be selected from a total of seven positions from b ₀ to b ₆ .

[00079] For example, the range of surrounding positions may be divided equally into N parts as described above. In practice, the unequal division method may also be adopted to divide the range of surrounding positions into N parts rather than into N equal parts, which is not limited to the present invention.

[00080] For example, after all the surrounding positions are equally divided into eight parts, seven weighted prediction positions can be obtained. In step 401, the encoder side can obtain one weighted prediction position from the seven weighted prediction positions, or select some weighted prediction positions (for example, five weighted prediction positions) from the seven weighted prediction positions, and then obtain one weighted prediction position from the five weighted prediction positions.

[00081] For example, the encoder side obtains the weighted prediction angle and the weighted prediction position of the current block in the following ways.

[00082] In the first method, the encoder side and the decoder side agree to accept the same weighted prediction angle as the weighted prediction angle of the current block, and agree to accept the same weighted prediction position as the weighted prediction position of the current block. For example, both the encoder side and the decoder side accept the weighted prediction angle A as the weighted prediction angle of the current block, and both sides accept the weighted prediction position a ₄ as the weighted prediction position of the current block.

[00083] In the second method, the encoder side creates a list of weighted prediction angles, which may include at least one weighted prediction angle, for example, the list of weighted prediction angles may include a weighted prediction angle A and a weighted prediction angle B. The encoder side creates a list of weighted prediction positions, which may include at least one weighted prediction position, for example, the list of weighted prediction positions may include weighted prediction positions ao-ae, weighted prediction positions b ₀ -b ₆ , etc. The encoder side sequentially goes through each weighted prediction angle in the list of weighted prediction angles and sequentially goes through each weighted prediction position in the list of weighted prediction positions, that is, goes through a combination of each weighted prediction angle and each weighted prediction position. For each combination of the weighted prediction angle(s) and the weighted prediction position(s), the weighted prediction angle and the weighted prediction position in the combination may be taken as the weighted prediction angle and the weighted prediction position of the current block, which are obtained in step 401, and steps 402-407 are performed based on the weighted prediction angle and the weighted prediction position to obtain a weighted prediction value of the current block.

[00084] For example, when the encoder side moves to the weighted prediction angle A and the weighted prediction position a ₀ , the encoder side executes steps 402-407 based on the weighted prediction angle A and the weighted prediction position ao to obtain the weighted prediction value A-1 of the current block. When the encoder side moves to the weighted prediction angle A and the weighted prediction position a ₁ , steps 402-407 are executed based on the weighted prediction angle A and the weighted prediction position a ₁ to obtain the weighted prediction value A-2 of the current block. When the encoder side obtains the weighted prediction angle B and the weighted prediction position b ₀ by traversal, steps 402-407 are executed based on the weighted prediction angle B and the weighted prediction position b ₀ to obtain the weighted prediction value B-1 of the current block, and so on. The encoder side can obtain the corresponding weighted prediction value based on each combination (each combination of weighted prediction angles and weighted prediction positions).

[00085] After obtaining all the weighted prediction values based on the combination of the weighted prediction angle(s) and the weighted prediction position(s), the encoder side may determine the RDO cost value based on each weighted prediction value in any unlimited manner. The encoder side may obtain the RDO cost value for each combination and select the minimum RDO cost value from all the RDO cost values.

[00086] Then, the encoder side takes the weighted prediction angle and the weighted prediction position in the combination corresponding to the minimum RDO cost value as the target weighted prediction angle and the target weighted prediction position of the current block, respectively, and finally encodes the index value of the target weighted prediction angle in the list of weighted prediction angles and the index value of the target weighted prediction position in the list of weighted prediction positions into a bitstream. A specific method for implementing the encoding of index values into a bitstream may refer to the application scenario discussed in the description of the second implementation method of the following embodiment of the invention.

[00087] The above-described methods are only illustrative and are not limited here, provided that the weighted prediction angle and the weighted prediction position of the current block can be obtained. For example, one weighted prediction angle can be selected in any way from the list of weighted prediction angles as the weighted prediction angle of the current block, and one weighted prediction position can be selected in any way from the list of weighted prediction positions as the weighted prediction position of the current block.

[00088] In step 402, for each pixel position of the current block, the encoder side determines a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block. For convenience of distinction, in this embodiment of the invention, the surrounding position outside the current block indicated by the pixel position may be referred to as a surrounding corresponding position for the pixel position.

[00089] For example, since a weighted prediction angle refers to an angular direction indicated by a pixel position within a current block, for each pixel position of the current block, the angular direction indicated by the pixel position is determined based on the weighted prediction angle, and then a surrounding corresponding position indicated by the pixel position is determined from surrounding positions outside the current block to which the pixel position belongs based on the angular direction.

[00090] For example, the surrounding positions outside the current block may include surrounding positions in the row above the current block, such as surrounding positions in the n1th row above the current block, where n1 may be equal to 1 or 2, 3, etc., which is not limited here; or may include surrounding positions in the column to the left of the current block, such as surrounding positions in the n2nd column to the left of the current block, where n2 may be 1 or 2, 3, etc., which is not limited here; or may include surrounding positions in the row below the current block, such as surrounding positions in the n3th row below the current block, where n3 may be equal to 1, 2, 3, etc., which is not limited here; or may include surrounding positions in the column to the right of the current block, such as surrounding positions in the n4th column to the right of the current block, where n4 may be 1 or 2, 3, etc., which is not limited here.

[00091] The above examples of surrounding positions are illustrative and are not limited here. In practice, in addition to the surrounding positions outside the current block, internal positions of the current block may also be used, that is, internal positions of the current block are used to replace the above-mentioned surrounding positions outside the current block, for example, internal positions may include internal positions in the n5th row inside the current block, where n5 may be 1 or 2, 3, etc., or internal positions in the n6th column inside the current block, where n6 may be 1 or 2, 3, etc. The length of internal positions may exceed the range of the current block, for example, the positions of the n7th row may exceed the range of the current block, that is, extend outside from two sides of the current block.

[00092] Internal positions of the current block and surrounding positions outside the current block may be used simultaneously.

[00093] To use internal positions of the current block or to use both internal positions of the current block and surrounding positions outside the current block, the current block may be divided into two sub-blocks, upper and lower, based on the row in which the internal positions are located, or divided into two sub-blocks, left and right, depending on the column in which the internal positions are located. In this case, the two sub-blocks have the same weighted prediction angle and the same weighted prediction position.

[00094] For example, surrounding positions beyond the current block may be located between pixel positions, such as at one or more sub-pixel positions, in which case the position of the current block cannot be simply described as a row "x", but is described as a row of sub-pixel positions located between the row "x" and the row "y".

[00095] For convenience of description, in the following embodiments of the invention, the surrounding positions in the first row from the top of the current block or the surrounding positions in the first column to the left of the current block are taken as an example, and other surrounding positions can be implemented in a similar manner.

[00096] For example, for a range of surrounding positions outside the current block, this range may be previously designated as a series of surrounding positions outside the current block. Alternatively, the range of surrounding positions outside the current block may be determined based on a weighted prediction angle, for example, the surrounding position indicated by each of the pixel positions within the current block is determined based on a weighted prediction angle, and then the range of surrounding positions outside the current block is determined based on the boundary of the surrounding positions indicated by all pixel positions. The range of surrounding positions is not limited here.

[00097] The surrounding positions outside the current block may include one or more integer pixel positions or one or more non-integer pixel positions. A non-integer pixel position may be a sub-pixel position, such as a 1/2 sub-pixel position, or a 1/4 sub-pixel position, or a 3/4 sub-pixel position, etc., which is not limited here. In addition, the surrounding positions outside the current block may include both one or more integer pixel positions and one or more sub-pixel positions.

[00098] For example, two surrounding positions outside the current block may correspond to one whole pixel position; or four surrounding positions outside the current block may correspond to one whole pixel position; or one surrounding position outside the current block may correspond to one whole pixel position; or one surrounding position outside the current block may correspond to two whole pixel positions. The above are just a few examples that do not limit the invention. The relationship between the surrounding position and the whole pixel position may be arbitrarily configured.

[00099] As shown in Fig. 4B and 4C, one surrounding position corresponds to one whole pixel position, and as shown in Fig. 4E, two surrounding positions correspond to one whole pixel position. Other examples will not be described in this embodiment for brevity.

[000100] At step 403, the encoder side determines a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position.

[000101] For example, for each pixel position of the current block, after determining the surrounding corresponding position indicated by the pixel position, the encoder side determines a reference weight value associated with the surrounding corresponding position. The reference weight value associated with the surrounding corresponding position may be pre-configured or determined based on a strategy, and a specific determination method may relate to subsequent embodiments of the invention.

[000102] Then, the encoder side determines a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position, for example, the reference weight value associated with the surrounding corresponding position may be determined as the target weight value of the pixel position.

[000103] For example, one or more reference weight values may be set for surrounding positions outside the current block. The specific process may be referred to in subsequent embodiments of the invention. The surrounding positions outside the current block may be one or more integer pixel positions or one or more subpixel positions. For example, one or more reference weight values may be set for one or more integer pixel positions outside the current block, and/or one or more reference weight values may be set for one or more subpixel positions outside the current block.

[000104] In a possible implementation, the target weight value of the pixel position is determined based on a reference weight value associated with the surrounding corresponding position, which may include the following five cases. Case 1. If the surrounding corresponding position is an integer pixel position, and the integer pixel position is set with a reference weight value, the target weight value of the pixel position may be determined based on the reference weight value of the integer pixel position. Case 2. If the surrounding corresponding position is an integer pixel position, and the integer pixel position is not set with a reference weight value, the target weight value of the pixel position may be determined based on the reference weight value(s) of the neighboring position(s) of the integer pixel position. For example, a rounding up operation is performed on the reference weight value of the neighboring position to obtain the target weight value of the pixel position; or performing a rounding down operation on a reference weight value of a neighboring position to obtain a target weight value of a pixel position; or determining the target weight value of a pixel position based on interpolation of reference weight values of neighboring positions of an integer pixel position. The determination method is not limited here. Case 3. If the surrounding corresponding position is a sub-pixel position, and the sub-pixel position is set with a reference weight value, the target weight value of a pixel position is determined based on the reference weight value of the sub-pixel position. Case 4. If the surrounding corresponding position is a sub-pixel position, and the sub-pixel position is not set with a reference weight value, the target weight value of a pixel position is determined based on one or more reference weight values of one or more neighboring positions of the sub-pixel position. For example, performing a rounding up operation on a reference weight value of a neighboring position to obtain a target weight value of a pixel position; or performing a rounding down operation on a reference weight value of a neighboring position to obtain a target weight value of a pixel position; or determining the target weight value of a pixel position based on interpolation of reference weight values of neighboring positions of a sub-pixel position. The determination method here is not limited. Case 5, determining the target weight value of a pixel position based on multiple reference weight values associated with a surrounding corresponding position. For example, regardless of whether the surrounding corresponding position is a whole pixel position or a sub-pixel position, reference weight values of a plurality of neighboring positions of the surrounding corresponding position can be obtained. If the surrounding corresponding position is set with a reference weight value, performing a weighting operation on the reference weight value of the surrounding corresponding position and the reference weight values of a plurality of neighboring positions to obtain the target weight value of a pixel position. If the surrounding corresponding position is not specified with a reference weight value, a weighting operation is performed on the reference weight values of a plurality of neighboring positions to obtain a target weight value of the pixel position.

[000105] The above cases 1-5 are only examples, and the methods for determining the target weight value are not limited here.

[000106] At step 404, the encoder side determines an associated pixel position weight value based on a target pixel position weight value.

[000107] At step 405, for each pixel position of the current block, the encoder side determines a first prediction value of the pixel position based on a first prediction mode of the current block and determines a second prediction value of the pixel position based on a second prediction mode of the current block.

[000108] For example, the first prediction mode may be any one of an intra-block copy prediction mode, an intra-prediction mode, an inter-prediction mode, and a palette mode; and the second prediction mode may be any one of an intra-block copy prediction mode, an intra-prediction mode, an inter-prediction mode, and a palette mode. For example, the first prediction mode may be an intra-block copy prediction mode, and the second prediction mode may be an intra-block copy prediction mode; the first prediction mode may be an inter-prediction mode, and the second prediction mode may be an inter-prediction mode, etc.

[000109] The process of determining a prediction value based on the first prediction mode and the second prediction mode may relate to subsequent embodiments of the invention.

[000110] At step 406, the encoder side determines a weighted pixel position prediction value based on the first pixel position prediction value, the target pixel position weight value, the second pixel position prediction value, and the associated pixel position weight value.

[000111] At step 407, the encoder side determines the weighted prediction values of the current block based on the weighted prediction value of each of the pixel positions of the current block.

[000112] As described above, in the embodiments of the present invention, an efficient method for setting weight values is provided for setting a suitable target weight value for each of the pixel positions of the current block, so that the prediction values of the current block are closer to the original pixels, which improves the prediction accuracy, the prediction efficiency, and the coding efficiency.

[000113] Embodiment 3 of the invention. In embodiment 1, the current block may include one sub-block, i.e., the sub-block is the current block itself. For the current block, Fig. 4F shows a schematic diagram illustrating a block diagram of a decoding method. The method is applied on the decoder side and includes the following steps 411-417.

[000114] In step 411, when it is determined that weighted prediction should be enabled for the current block, the decoder side obtains a weighted prediction angle and a weighted prediction position of the current block.

[000115] For example, the decoder side also needs to determine whether to enable weighted prediction for the current block. If the decoder side determines that weighted prediction should be enabled for the current block, the weighted prediction angle and the weighted prediction position of the current block are obtained and subsequent steps are performed, otherwise the processing method is not limited in the present invention.

[000116] In a possible implementation, the encoder side determines whether the characteristic information of the current block satisfies a certain condition, and if the characteristic information satisfies the certain condition, the encoder side determines that weighted prediction should be enabled for the current block. The decoder side also determines whether the characteristic information of the current block satisfies a certain condition, and if the characteristic information satisfies the certain condition, the decoder side determines that weighted prediction should be enabled for the current block, otherwise the decoder side determines that weighted prediction should not be enabled for the current block. The manner in which the decoder side determines whether to enable weighted prediction for the current block based on the characteristic information may be referred to step 401 and will not be repeated here.

[000117] In another possible implementation, on the encoder side, it is determined whether the current block supports weighted prediction based on the characteristic information of the current block. When the current block supports weighted prediction, another strategy can also be adopted to determine whether to enable weighted prediction for the current block, for example, using RDO to determine whether to enable weighted prediction for the current block. After determining whether to enable weighted prediction for the current block, when the encoder side transmits the coded bitstream of the current block, the coded bitstream can include syntax for determining whether to enable weighted prediction for the current block, wherein the syntax indicates whether to enable weighted prediction for the current block. The decoder side determines whether the current block supports weighted prediction based on the characteristic information of the current block, and the specific implementation can be referred to step 401 and is not repeated here. When it is determined that the current block supports weighted prediction, the decoder side may also decode syntax for determining whether to enable weighted prediction for the current block from the encoded bitstream and determine whether to enable weighted prediction for the current block based on the syntax.

[000118] In a possible implementation, when it is determined that weighted prediction should be enabled for the current block, the decoder side obtains a weighted prediction angle and a weighted prediction position of the current block, wherein the corresponding description of the weighted prediction angle and the weighted prediction position may refer to step 401 and will not be repeated here. The decoder side may obtain the weighted prediction angle and the weighted prediction position of the current block in the following ways.

[000119] In the first method, the decoder side and the encoder side agree to adopt the same weighted prediction angle as the weighted prediction angle of the current block, and agree to adopt the same weighted prediction position as the weighted prediction position of the current block. Specific content may relate to the first method in the embodiment of the invention described above.

[000120] In the second method, a list of weighted prediction angles is created on the decoder side, which is the same as the list of weighted prediction angles on the encoder side. A list of weighted prediction positions is created on the decoder side, which is the same as the list of weighted prediction positions on the encoder side. Specific content may relate to the second method in the embodiment of the invention described above. After obtaining the coded bitstream of the current block, the decoder side analyzes the indication information from the coded bitstream and selects one weighted prediction angle from the list of weighted prediction angles as the weighted prediction angle of the current block based on the indication information, and selects one weighted prediction position from the list of weighted prediction positions as the weighted prediction position of the current block based on the indication information.

[000121] The process of implementing the second method will be described below in conjunction with several specific application scenarios.

[000122] Application scenario 1: When the encoder side transmits an encoded bitstream to the decoder side, the encoded bitstream may include indication information 1 used to indicate both a weighted prediction angle of the current block (that is, a target weighted prediction angle) and a weighted prediction position of the current block (that is, a target weighted prediction position). For example, when indication information 1 is 0, the indication information 1 may indicate the first weighted prediction angle in the list of weighted prediction angles and the first weighted prediction position in the list of weighted prediction positions, etc. There is no limitation on the meaning of the indication information 1 to indicate which weighted prediction angle and weighted prediction position are used, if the encoder side and the decoder side agree thereon.

[000123] After receiving the coded bitstream, the decoder side decodes the indication information 1 from the coded bitstream. Based on the indication information 1, the decoder side may select a weighted prediction angle corresponding to the indication information 1 from a list of weighted prediction angles, and adopt the weighted prediction angle as a weighted prediction angle of the current block. Based on the indication information 1, the decoder side may select a weighted prediction position corresponding to the indication information 1 from a list of weighted prediction positions, and adopt the weighted prediction position as a weighted prediction position of the current block.

[000124] Application scenario 2: When the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may include indication information 2 and indication information 3. The indication information 2 indicates a target weighted prediction angle of the current block, for example, an index value 1 of the target weighted prediction angle in the weighted prediction angle list, wherein the index value 1 indicates which weighted prediction angle in the weighted prediction angle list is the target weighted prediction angle. The indication information 3 indicates a target weighted prediction position of the current block, for example, an index value 2 of the target weighted prediction position in the weighted prediction position list, wherein the index value 2 indicates which weighted prediction position in the weighted prediction position list is the target weighted prediction position.

[000125] After receiving the coded bitstream, the decoder side analyzes the indication information 2 and the indication information 3 from the coded bitstream. Based on the indication information 2, the decoder side can select a weighted prediction angle corresponding to the index value 1 in the weighted prediction angle list, and take the weighted prediction angle as the weighted prediction angle of the current block. Based on the indication information 3, the decoder side can select a weighted prediction position corresponding to the index value 2 in the weighted prediction position list, and take the weighted prediction position as the weighted prediction position of the current block.

[000126] Application scenario 3: the encoder side and the decoder side may agree on a preferred configuration combination. The preferred configuration combination may be configured according to experience, which is not limited here. For example, they may agree that the preferred configuration combination includes a preferred configuration combination 1 including a weighted prediction angle A and a weighted prediction position a ₄ , a preferred configuration combination 2 including a weighted prediction angle B and a weighted prediction position b ₄ , and the like.

[000127] After determining the target weighted prediction angle and the target weighted prediction position of the current block, the encoder side determines whether the combination of the target weighted prediction angle and the target weighted prediction position belongs to the preferred configuration combination. If the combination is included in the preferred configuration combination, when the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may include indication information 4 and indication information 5. The indication information 4 indicates whether the preferred configuration combination is used for the current block. For example, the indication information 4 is a first value (for example, 0) indicating that the preferred configuration combination is used for the current block. The indication information 5 indicates which preferred configuration combination is used for the current block. For example, when the value of the indication information 5 is 0, this may indicate that the preferred configuration combination 1 is used for the current block, and when the value of the indication information 5 is 1, this may indicate that the preferred configuration combination 2 is used for the current block.

[000128] After receiving the coded bitstream, the decoder side may analyze the indication information 4 and the indication information 5 from the coded bitstream. Based on the indication information 4, the decoder side can determine whether the preferred configuration combination is used for the current block. If the indication information 4 is the first value, the decoder side determines that the preferred configuration combination is used for the current block. When the preferred configuration combination is used for the current block, the decoder side can determine which preferred configuration combination is used for the current block based on the indication information 5. For example, when the value of the indication information 5 is 0, the decoder side can determine that the preferred configuration combination 1 is used for the current block, for example, the weighted prediction angle of the current block is the weighted prediction angle A, and the weighted prediction position of the current block is the weighted prediction position a ₄ . In another example, when the value of the indication information 5 is 1, the decoder side may determine that the preferred configuration combination 2 is used for the current block, for example, the weighted prediction angle of the current block is the weighted prediction angle B, and the weighted prediction position of the current block is the weighted prediction position b ₄ .

[000129] For example, if the encoder side and the decoder side agree on only one set of combinations in the preferred configuration combination, for example, the preferred configuration combination includes only the weighted prediction angle A and the weighted prediction position a ₄ , the coded bitstream may include only the indication information 4 without the indication information 5, wherein the indication information 4 indicates that the preferred configuration combination is used for the current block. After the decoder side analyzes the indication information 4 from the coded bitstream, if the indication information 4 is the first value, the decoder side determines that the preferred configuration combination is used for the current block. Based on the preferred combination, the decoder side determines that the weighted prediction angle of the current block is the weighted prediction angle A, and the weighted prediction position of the current block is the weighted prediction position a ₄ .

[000130] Application scenario 4: The encoder side and the decoder side may agree on a preferred configuration combination. After determining the target weighted prediction angle and the target weighted prediction position of the current block, the encoder side may determine whether the combination of the target weighted prediction angle and the target weighted prediction position belongs to the preferred configuration combination. If it does not belong to the preferred configuration combination, when the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may include indication information 4 and indication information 6. The indication information 4 indicates whether the preferred configuration combination is used for the current block. If the indication information 4 is the second value (for example, 1), the indication information 4 may indicate that the preferred configuration combination is not used for the current block. The indication information 6 indicates both the target weighted prediction angle of the current block and the target weighted prediction position of the current block.

[000131] After receiving the coded bitstream, the decoder side may analyze the indication information 4 and the indication information 6 from the coded bitstream. Based on the indication information 4, the decoder side may determine whether the preferred configuration combination is used for the current block. If the indication information 4 is the second value, the decoder side determines that the preferred configuration combination is not used for the current block. When the preferred configuration combination is not used for the current block, the decoder side may select, based on the indication information 6, a weighted prediction angle corresponding to the indication information 6 from the list of weighted prediction angles, and take the weighted prediction angle as the weighted prediction angle of the current block; and based on the indication information 6, the decoder side may select a weighted prediction position corresponding to the indication information 6 from the list of weighted prediction positions, and take the weighted prediction position as the weighted prediction position of the current block.

[000132] Application scenario 5: The encoder side and the decoder side may agree on a preferred configuration combination. After determining the target weighted prediction angle and the target weighted prediction position of the current block, the encoder side may determine whether the combination of the target weighted prediction angle and the target weighted prediction position is included in the preferred configuration combination. If the combination is not the preferred configuration combination, when the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may include indication information 4, indication information 7, and indication information 8. For example, indication information 4 indicates whether the preferred configuration combination is used for the current block. If indication information 4 is the second value, indication information 4 may indicate that the preferred configuration combination is not used for the current block. Indication information 7 indicates the target weighted prediction angle of the current block. Indication information 8 indicates the target weighted prediction position of the current block.

[000133] After receiving the coded bitstream, the decoder side analyzes the indication information 4, the indication information 7 and the indication information 8 from the coded bitstream. Based on the indication information 4, the decoder side can determine whether the preferred configuration combination is used for the current block. If the indication information 4 is the second value, the decoder side determines that the preferred configuration combination is not used for the current block. When the preferred configuration combination is not used for the current block, the decoder side can select, based on the indication information 7, a corresponding weighted prediction angle from the list of weighted prediction angles and take the weighted prediction angle as the weighted prediction angle of the current block; and based on the indication information 8, the decoder side can select a corresponding weighted prediction position from the list of weighted prediction positions and take the weighted prediction position as the weighted prediction position of the current block.

[000134] The above-mentioned first and second methods are merely illustrative and are not limited here, provided that the decoder side can obtain the weighted prediction angle (that is, the target weighted prediction angle) and the weighted prediction position (that is, the target weighted prediction position) of the current block.

[000135] At step 412, for each pixel position of the current block, the decoder side determines a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on a weighted prediction angle of the current block.

[000136] In step 413, the decoder side determines a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position.

[000137] At step 414, the decoder side determines an associated pixel position weight value based on the target pixel position weight value.

[000138] At step 415, for each pixel position of the current block, the decoder side determines a first prediction value of the pixel position based on a first prediction mode of the current block and determines a second prediction value of the pixel position based on a second prediction mode of the current block.

[000139] At step 416, based on the first prediction value, the target weight value, the second prediction value, and the associated pixel position weight value, the decoder side determines a weighted prediction value of the pixel position.

[000140] In step 417, the decoder side determines the weighted prediction values of the current block based on the weighted prediction value of each of the pixel positions of the current block.

[000141] For example, the processes of implementing steps 412-417 may be classified as steps 402-407, with the difference that steps 412-417 are decoder-side processes rather than encoder-side processes, and so will not be repeated here.

[000142] As described above, in the embodiments of the present invention, an efficient method for setting weight values is provided, in which a suitable target weight value is set for each of the pixel positions of the current block so that the prediction values of the current block are closer to the original pixels, which improves the prediction accuracy, the prediction efficiency, and the decoding efficiency.

[000143] In the above embodiments 2 and 3 of the invention, the current block includes one sub-block, that is, the processing of the current block can be designated as the angular weighted prediction (AWP) mode.

[000144] Embodiment 4. In Embodiments 1 to 3 of the invention above, a weighted prediction angle is used. The weighted prediction angle may be any angle, such as any angle within 180 degrees or any angle within 360 degrees, such as 10 degrees, 20 degrees, 30 degrees, etc., which is not limited here. In a possible implementation, the weighted prediction angle may be a horizontal angle (for example, angle 2 in Fig. 5B); or the weighted prediction angle may be a vertical angle (for example, angle 6 in Fig. 5B); or the absolute value of the slope of the weighted prediction angle (the slope of the weighted prediction angle is the tangent of the weighted prediction angle) may be the n-th power of 2, where n is an integer, such as a positive integer, 0, and a negative integer. For example, the absolute value of the slope of the weighted prediction angle may be 1 (that is, the value of the 0th power of 2), 2 (that is, the value of the 1st power of 2), 1/2 (that is, the value of - 1st power of 2), 4 (that is, the value of the 2nd power of 2), 1/4 (that is, the value of - 2nd power of 2), 8 (that is, the value of the 3rd power of 2), and 1/8 (that is, the value of - 3rd power of 2), etc. For example, Fig. 5A shows eight weighted prediction angles, where the absolute values of the slopes of the weighted prediction angles are the values of the nth power of 2. In the following embodiments of the invention, a shift operation may be performed on a weighted prediction angle. Therefore, when the absolute value of the slope of the weighted prediction angle is the nth power of 2, the division operation can be avoided during the shift operation for the weighted prediction angle, which facilitates the implementation of the shift.

[000145] For example, the number of weighted prediction angles supported by different block sizes may be the same or different.

[000146] Embodiment 5. In the above Embodiments 1 to 3, a target weight value of the pixel position may be determined for each pixel position based on a reference weight value associated with a surrounding corresponding position indicated by the pixel position. In Embodiments 2 and 3, the surrounding corresponding position is a surrounding corresponding position of the current block.

[000147] A reference weight value associated with a surrounding corresponding position may be obtained as follows: based on a coordinate value of a surrounding corresponding position of a current block and a coordinate value of a start position of a weight transformation of the current block, a reference weight value associated with a surrounding corresponding position is determined.

[000148] For example, if the surrounding corresponding position is a surrounding position in one row above the current block or in one row below the current block, the coordinate value of the surrounding corresponding position is the abscissa value of the surrounding corresponding position, and the coordinate value of the starting position of the weight transformation is the abscissa value of the starting position of the weight transformation. Alternatively, if the surrounding corresponding position is a surrounding position in one column to the left of the current block or in one column to the right of the current block, the coordinate value of the surrounding corresponding position is the ordinate value of the surrounding corresponding position, and the coordinate value of the starting position of the weight transformation is the ordinate value of the starting position of the weight transformation.

[000149] For example, a pixel position in the upper left corner of the current block (for example, the first pixel position in the upper left corner) can be used as the origin, and a coordinate value of the surrounding corresponding position of the current block (for example, an abscissa value or an ordinate value) and a coordinate value of the starting position of the weight transformation of the current block (for example, an abscissa value or an ordinate value) are both coordinate values relative to the origin. Another pixel position of the current block can also be taken as the origin, and the implementation method is similar to the method of using the pixel position in the upper left corner as the origin.

[000150] In a possible implementation, when a reference weight value associated with the surrounding corresponding position is determined based on the coordinate value of the surrounding corresponding position and the coordinate value of the initial position of the weight transformation of the current block, a difference between the coordinate value of the surrounding corresponding position and the coordinate value of the initial position of the weight transformation of the current block can be calculated. If the difference is less than a first threshold value, the reference weight value associated with the surrounding corresponding position is determined as the first threshold value. If the difference is greater than a second threshold value, the reference weight value associated with the surrounding corresponding position is determined as the second threshold value. If the difference is not less than the first threshold value and is not greater than the second threshold value, the reference weight value associated with the surrounding corresponding position is determined as the difference. In another possible implementation, when determining a reference weight value associated with a surrounding corresponding position based on a coordinate value of the surrounding corresponding position and a coordinate value of the initial position of the weight transformation of the current block, the reference weight value associated with the surrounding corresponding position is determined directly based on the size relationship between the coordinate value of the surrounding corresponding position and the coordinate value of the initial position of the weight transformation of the current block.

[000151] For example, if the coordinate value of the surrounding corresponding position is less than the coordinate value of the starting position of the weight transformation of the current block, the reference weight value associated with the surrounding corresponding position may be determined as a first threshold value; and if the coordinate value of the surrounding corresponding position is not less than the coordinate value of the starting position of the weight transformation of the current block, the reference weight value associated with the surrounding corresponding position may be determined as a second threshold value.

[000152] As another example, if the coordinate value of the surrounding corresponding position is less than the coordinate value of the starting position of the weight transformation of the current block, the reference weight value associated with the surrounding corresponding position may be determined as a second threshold value; and if the coordinate value of the surrounding corresponding position is not less than the coordinate value of the starting position of the weight transformation of the current block, the reference weight value associated with the surrounding corresponding position may be determined as a first threshold value.

[000153] For example, the first threshold value and the second threshold value may be configured based on experience, wherein the first threshold value is less than the second threshold value. The first threshold value and the second threshold value are not limited here. In some examples, the first threshold value may be a minimum value of the pre-agreed reference weight values, such as 0, and the second threshold value may be a maximum value of the pre-agreed reference weight values, such as 8, where 0 and 8 are only examples.

[000154] For example, as shown in Fig. 4D, all the surrounding positions are equally divided into eight parts to obtain seven weighted prediction positions. When the weighted prediction position is 0, it points to the surrounding position a ₀ , that is, the coordinate value of the starting position of the weight transformation is the coordinate value of the surrounding position a ₀ . When the weighted prediction position is 1, it points to the surrounding position a ₁ , that is, the coordinate value of the starting position of the weight transformation is the coordinate value of the surrounding position a ₁ , and so on. The method for determining the coordinate value of the starting position of the weight transformation will not be repeated here.

[000155] Embodiment 6. In Embodiments 1 to 3 above, for each pixel position, a target weight value of the pixel position may be determined based on a reference weight value associated with a surrounding corresponding position indicated by the pixel position. The following description is given for a surrounding corresponding position of the current block as an example. The reference weight value associated with the surrounding corresponding position may be obtained as follows: a list of reference weight values of the current block is determined, wherein the list of reference weight values may include a plurality of reference weight values that are preconfigured or configured based on weight configuration parameters. Based on a target index, an actual number of reference weight values is selected from the list of reference weight values, and one or more reference weight values are set for surrounding positions outside the current block based on the actual number of reference weight values. For example, the actual number may be determined based on the size of the current block and the weighted prediction angle of the current block; The target index can be determined based on the size of the current block, the weighted prediction angle of the current block, and the weighted prediction position of the current block.

[000156] Thus, since one or more reference weight values have already been set for the surrounding positions outside the current block, that is, each of the surrounding positions has a reference weight value, after determining the surrounding corresponding position indicated by the pixel position, a reference weight value associated with the surrounding corresponding position, for example, a target weight value of the pixel position, can be determined from the surrounding positions outside the current block.

[000157] In combination with the specific implementation steps shown in Fig. 6A, the process of setting the reference weight values of the surrounding positions will be described below.

[000158] At step S1, a list of reference weight values of the current block is determined.

[000159] In a possible implementation, the list of reference weight values at the sequence level may be defined as the list of reference weight values of the current block. For example, both the encoder side and the decoder side may be configured with a list A1 of reference weight values of the sequence level. For pictures with multiple sequence levels, all of these pictures use the list A1 of reference weight values. Regardless of what the weighted prediction angle and the weighted prediction position are, the current blocks of a plurality of pictures of the sequence level share the same list A1 of reference weight values.

[000160] In another possible implementation, the preset list of reference weight values may be defined as a list of reference weight values of the current block. For example, both the encoder side and the decoder side may preset a list of reference weight values. This list of reference weight values is used for all pictures in several sequences. Regardless of the weighted prediction angle and the weighted prediction position, the current blocks of all pictures in the plurality of sequences share the same list of reference weight values.

[000161] In another possible implementation, a list of reference weight values corresponding to a weighted prediction angle of the current block may be defined as a list of reference weight values of the current block. For example, both the encoder side and the decoder side are configured with a plurality of lists of reference weight values, and several weighted prediction angles may share the same list of reference weight values. For example, a list of reference weight values A2 and a list of reference weight values A3 are configured, weighted prediction angle 1 and weighted prediction angle 2 share the list of reference weight values A2, and weighted prediction angle 3 uses the list of reference weight values A3. Based on this, after obtaining the weighted prediction angle of the current block, if the weighted prediction angle of the current block is weighted prediction angle 1, the list of reference weight values A2 corresponding to weighted prediction angle 1 is defined as a list of reference weight values of the current block.

[000162] In another possible implementation, a list of reference weight values corresponding to a weighted prediction angle of the current block and a weighted prediction position of the current block may be defined as a list of reference weight values of the current block. For example, both the encoder side and the decoder side are configured with a plurality of lists of reference weight values, different combinations of a weighted prediction angle and a weighted prediction position may correspond to the same or different lists of reference weight values. For example, a list of reference weight values A4, a list of reference weight values A5, and a list of reference weight values A6 are configured. Weighted prediction angle 1 and weighted prediction positions 0 to 2 share the list of reference weight values A4, weighted prediction angle 1 and weighted prediction positions 3 to 5 share the list of reference weight values A5, and weighted prediction angle 2 and weighted prediction positions 0 to 5 share the list of reference weight values A6. Based on this, after obtaining the weighted prediction angle and the weighted prediction position of the current block, if the weighted prediction angle of the current block is weighted prediction angle 1 and the weighted prediction position of the current block is weighted prediction position 4, the list A5 of reference weight values corresponding to weighted prediction angle 1 and weighted prediction position 4 is defined as the list of reference weight values of the current block.

[000163] In another possible implementation, a list of reference weight values corresponding to the size of the current block and the weighted prediction angle of the current block is defined as a list of reference weight values of the current block. For example, both the encoder side and the decoder side are configured with a plurality of lists of reference weight values, different combinations of the size and the weighted prediction angle may correspond to the same or different lists of reference weight values. For example, a list of reference weight values A7 and a list of reference weight values A8 are configured. The weighted prediction angle 1 and the size 1 use the list of reference weight values A7. The weighted prediction angle 1 and the size 2, as well as the weighted prediction angle 2 and the size 1 use the list of reference weight values A8. Based on this, if the weighted prediction angle of the current block is weighted prediction angle 1 and the size of the current block is size 1, a list A7 of reference weight values corresponding to weighted prediction angle 1 and size 1 is determined as a list of reference weight values of the current block.

[000164] In this manner, a list of reference weight values of the current block may be defined, and the list of reference weight values may include a plurality of reference weight values that are pre-configured or configured based on the weight configuration parameters.

[000165] For the list of reference weight values of the current block, the number of reference weight values in the list of reference weight values may be a predetermined fixed value, and the fixed value may be set arbitrarily based on experience, which is not limited here. Alternatively, the number of reference weight values in the list of reference weight values may be related to the size (for example, width or height) of the current slice in which the current block is located, for example, the number of reference weight values may be greater than the width of the current slice or the same as the width of the current slice; and the number of reference weight values may be greater than the height of the current slice or the same as the height of the current slice, which is not limited here. The number of reference weight values may be selected based on practical needs.

[000166] For example, for a plurality of reference weight values in a list of reference weight values, the plurality of reference weight values may be non-consecutive reference weight values, such that the plurality of reference weight values in the list of reference weight values may not be completely the same.

[000167] In a possible implementation, the plurality of reference weight values in the list of reference weight values may monotonically increase or monotonically decrease. In addition, the plurality of reference weight values in the list of reference weight values may first monotonically increase and then monotonically decrease, or first monotonically decrease and then monotonically increase. In addition, the plurality of reference weight values in the list of reference weight values may first include a plurality of first reference weight values and then a plurality of second reference weight values, or first include a plurality of second reference weight values and then a plurality of first reference weight values. In combination with several cases, the content of the above list of reference weight values is described below.

[000168] Case 1: The plurality of reference weight values in the reference weight value list may monotonically increase or monotonically decrease. For example, the reference weight value list may be [8 8 8 8…8 8 7 6 5 4 3 2 1 0 0 0 0…0 0], that is, the plurality of reference weight values in the reference weight value list monotonically decrease. In another example, the reference weight value list may be [0 0 0 0…0 0 1 2 3 4 5 6 7 8 8 8 8…8 8], that is, the plurality of reference weight values in the reference weight value list monotonically increase. The above are only examples, and the list of reference weight values is not limited here.

[000169] For example, the reference weight values in the reference weight value list may be pre-configured or configured based on weight configuration parameters. The weight configuration parameters may include a weight conversion factor and a weight conversion start position. The weight conversion factor may be a value set based on experience, and the weight conversion start position may also be a value set based on experience.

[000170] The plurality of reference weight values in the list of reference weight values may monotonically increase or monotonically decrease. For example, if the maximum value of the reference weight values is M1 and the minimum value of the reference weight values is M2, the plurality of reference weight values in the list of reference weight values monotonically decrease from the maximum value of M1 to the minimum value of M2; or monotonically increase from the minimum value of M2 to the maximum value of M1. If M1 is 8 and M2 is 0, the plurality of reference weight values monotonically decrease from 8 to 0 or monotonically increase from 0 to 8.

[000171] For example, in a process of pre-configuring reference weight values in a reference weight value list, a plurality of reference weight values in the reference weight value list may be arbitrarily configured under the condition that the plurality of reference weight values monotonically increase or monotonically decrease.

[000172] For example, in the process of configuring reference weight values in the reference weight value list based on the weight configuration parameters, a weight conversion factor and a weight conversion start position may first be obtained, and then a plurality of reference weight values in the reference weight value list are determined based on the weight conversion factor and the weight conversion start position. The weight conversion factor and the weight conversion start position may be preset values that can be configured based on experience and are not limited here.

[000173] For example, the reference weight values in the reference weight value list may be defined as follows: y(x)=Clip3(minimum, maximum, a*(x-s)), where x represents a position index in the reference weight value list, for example, x is 1, which indicates the first position in the reference weight value list, y(x) represents the x-th reference weight value in the reference weight value list, "a" represents a weight conversion coefficient, s represents a weight conversion start position, and the Clip3 function is used to limit the reference weight values between the minimum and maximum values. The maximum value and the minimum value may be configured based on experience. For simplicity, the following description is given for a minimum value of 0 and a maximum value of 8 as an example.

[000174] "a" is a weight conversion factor that can be configured based on experience, for example, "a" can be a non-zero integer, for example, "a" can be -4, -3, -2, -1, 1, 2, 3 or 4, etc. For simplicity, the description is provided for "a" equal to 1 as an example. If the factor "a" is 1, the reference weight values should pass through 0, 1, 2, 3, 4, 5, 6, 7, 8 from 0 to 8 or pass through 8, 7, 6, 5, 4, 3, 2, 1, 0 from 8 to 0.

[000175] s may represent a starting position of the weight conversion, which may be configured based on experience, for example, s may be half of the total number of reference weight values in the list of reference weight values; or s may be slightly less than half of the total number of reference weight values, for example, half of the total number of reference weight values minus 4; or s may be slightly greater than half of the total number of reference weight values, for example, half of the total number of reference weight values plus 4. The above values of s are only examples and are not limited here.

[000176] To summarize, the reference weight values in the reference weight value list can be configured based on the weight configuration parameters as follows: ReferenceWeightsWhole[x] = Clip3(0, 8, x - Z); or ReferenceWeightsWhole[x] = Clip3(0, 8, Z - x); or ReferenceWeightsWhole[x] = Clip3(0, 4, x - Z); or ReferenceWeightsWhole[x] = Clip3(0, 4, Z - x). The above-mentioned methods are only examples, and the implementation method is not limited here.

[000177] In the above formulas, x is in the range from 0 to WholeLength-1. When x is 1, ReferenceWeightsWhole[x] may represent the first reference weight value in the list of reference weight values; when x is 2, ReferenceWeightsWhole[x] may represent the second reference weight value in the list of reference weight values, and so on. For example, if the surrounding positions beyond the current block are surrounding positions in the same top row or in the same bottom row, WholeLength is determined based on the width of the current slice; and if the surrounding positions beyond the current block are surrounding positions in the left column or the right column, WholeLength is determined based on the height of the current slice.

[000178] In the above formula a*(x-s), if the value of "a" is 1, a*(x-s)=x-s, that is, x-Z in Clip3(0, 8, x - Z) is equivalent to x-s, where Z represents the starting position of the weight transformation. If "a" is -1, a*(x-s)=s-x, that is, Z-x in Clip3(0, 8, Z - x) is equivalent to s-x, where Z represents the starting position of the weight transformation. When "a" is another value, the implementation process is similar, provided that the reference weight values in the reference weight value list satisfy y(x)=Clip3(minimum, maximum, a*(x-s)). Clip3(0, 8) is used to restrict the reference weight values from 0 to 8, and Clip3(0, 4) is used to restrict the reference weight values from 0 to 4.

[000179] In the above formulas, Z represents the initial position of the weight transformation, which can be configured based on experience. Assuming that x is in the range from 0 to 511, and Z is 255, Z is substituted into the formula ReferenceWeightsWhole[x]=Clip3(0, 8, x - Z), and for any value from 0 to 511, ReferenceWeightsWhole[x] can be obtained, that is, 512 reference weight values can be obtained, and the 512 reference weight values form a list of reference weight values. For example, when x is in the range from 0 to 255, all the corresponding reference weight values are 0, when x is 256, the reference weight value is 1, and so on, when x is 262, the reference weight value is 7, and when x varies from 263 to 511, all the reference weight values are 8. In short, when the value of "a" is 1, Clip3(0, 8, x - Z) is used to monotonically increase the reference weight values. Similarly, Z can be substituted into another formula to obtain 512 reference weight values, and the list of reference weight values is formed based on the 512 reference weight values. For example, when the value of "a" is -1, Clip3(0, 8, Z - x) is used to monotonically decrease the reference weight values. Clip3(0, 4) is used to limit the reference weight values from 0 to 4, which is not repeated here.

[000180] In short, a list of reference weight values of the current block may be obtained, and the list of reference weight values may include a plurality of reference weight values that may monotonically increase or monotonically decrease. In a possible implementation, the list of reference weight values may also include one or more reference weight values of target positions (areas), one or more reference weight values of first neighboring positions (areas) of the target positions, and one or more reference weight values of second neighboring positions (areas) of the target positions.

[000181] The target positions include one or more reference weight values determined based on the initial position of the weight transformation. For example, based on the initial position of the weight transformation, one reference weight value is determined and taken as the target positions. For example, the initial position s of the weight transformation is 255, the 259th reference weight value can be taken as the target positions, or the 258th reference weight value can be taken as the target positions, or the 260th reference weight value can be taken as the target positions. The above are just a few examples and are not limited here. In another example, based on the initial position of the weight transformation, a plurality of reference weight values are determined and taken as the target positions. For example, the 256th to 262nd reference weight values are taken as the target positions, or the 258th to 260th reference weight values are taken as the target positions. The above are just a few examples and are not limited here.

[000182] For example, the target positions may include a reference weight value with a value of 4. For example, the 259th reference weight value is 4. Therefore, if the target positions include one reference weight value, the target positions may include the 259th reference weight value, or if the target positions include a plurality of reference weight values, the target positions may include the 256th to 262nd reference weight values or the 258th to 260th reference weight values, which is not limited here, provided that the 259th reference weight value is within the target positions.

[000183] In short, the target positions may include one reference weight value; or the target positions may include a plurality of reference weight values. If the target positions include a plurality of reference weight values, the plurality of reference weight values of the target positions monotonically increase or monotonically decrease. The monotonous increase may be a strictly monotonous increase (that is, the plurality of reference weight values at the target positions strictly monotonically increase); and the monotonous decrease may be a strictly monotonous decrease (that is, the plurality of reference weight values at the target positions strictly monotonically decrease). For example, the plurality of reference weight values of the target positions monotonically increase from 1 to 7 or monotonically decrease from 7 to 1.

[000184] For example, all the reference weight values of the first neighboring positions are the first reference weight values, and the reference weight values of the second neighboring positions monotonically increase or monotonically decrease. For example, all the reference weight values of the first neighboring positions are equal to 0, the target positions include one reference weight value equal to 1, and the reference weight values of the second neighboring positions monotonically increase from 2 to 8.

[000185] Optionally, all the reference weight values of the first adjacent positions are second reference weight values, and all the reference weight values of the second adjacent positions are third reference weight values, wherein the second reference weight values are different from the third reference weight values. For example, all the reference weight values of the first adjacent positions are 0, the target positions include a plurality of reference weight values that monotonically increase from 1 to 7, and the reference weight values of the second adjacent positions are 8. Obviously, the reference weight values of the first adjacent positions are different from the values of the second adjacent positions.

[000186] Optionally, the reference weight values of the first neighboring positions and the reference weight values of the second neighboring positions monotonically increase or monotonically decrease simultaneously. For example, the reference weight values of the first neighboring positions monotonically increase, and the reference weight values of the second neighboring positions also monotonically increase. In another example, the reference weight values of the first neighboring positions monotonically decrease, and the reference weight values of the second neighboring positions also monotonically decrease. For example, the reference weight values of the first neighboring positions monotonically increase from 0 to 3, the target positions include one reference weight value equal to 4, and the reference weight values of the second neighboring positions monotonically increase from 5 to 8.

[000187] Case 2: a plurality of reference weight values in the reference weight value list first monotonically increase and then monotonically decrease, or first monotonically decrease and then monotonically increase. For example, the reference weight value list may be [8 8…8 8 7 6 5 4 3 2 1 0 0…0 0 1 2 3 4 5 6 7 8 8…8 8], that is, a plurality of reference weight values in the reference weight value list first monotonically decrease and then monotonically increase. In another example, the reference weight value list may be [0 0…0 0 1 2 3 4 5 6 7 8 8…8 8 7 6 5 4 3 2 1 0 0…0 0], that is, a plurality of reference weight values in the reference weight value list first monotonically increase and then monotonically decrease. The above examples are illustrative only and the list of reference weight values is not limited here.

[000188] For example, the reference weight values in the reference weight value list may be pre-configured or configured based on weight configuration parameters. The weight configuration parameters may include a weight conversion factor and a weight conversion start position. The weight conversion factor may be a value set based on experience, and the weight conversion start position may also be a value set based on experience.

[000189] For example, if the maximum value of the reference weight values is M1 and the minimum value of the reference weight values is M2, the plurality of reference weight values in the list of reference weight values monotonically decrease from the maximum value of M1 to the minimum value of M2 and then monotonically increase from the minimum value of M2 to the maximum value of M1, or monotonically increase from the minimum value of M2 to the maximum value of M1 and then monotonically decrease from the maximum value of M1 to the minimum value of M2. If M1 is 8 and M2 is 0, the plurality of reference weight values monotonically decrease from 8 to 0 and then monotonically increase from 0 to 8; or the plurality of reference weight values monotonically increase from 0 to 8 and then monotonically decrease from 8 to 0.

[000190] For example, in the process of pre-configuring the reference weight values in the reference weight value list, a plurality of reference weight values in the reference weight value list may be arbitrarily configured, provided that the plurality of reference weight values first monotonically increase and then monotonically decrease, or the plurality of reference weight values first monotonically decrease and then monotonically increase. There are no other restrictions on the reference weight values.

[000191] For example, in the process of configuring reference weight values in the reference weight value list based on the weight configuration parameters, a first weight conversion factor, a second weight conversion factor, a first weight conversion start position, and a second weight conversion start position may first be obtained, and then a plurality of reference weight values in the reference weight value list are determined based on the first weight conversion factor, the second weight conversion factor, the first weight conversion start position, and the second weight conversion start position. The first weight conversion factor, the second weight conversion factor, the first weight conversion start position, and the second weight conversion start position may be predetermined values, and no restrictions are imposed on the first weight conversion factor, the second weight conversion factor, the first weight conversion start position, and the second weight conversion start position.

[000192] For example, the reference weight values in the reference weight value list may be defined as follows: when x is in [0, k], y(x)=Clip3(minimum, maximum, a1*(x-s1)). When x is in [k+1, t], y(x)=Clip3(minimum, maximum, a2*(x-s2)). x represents a position index in the reference weight value list, for example, x is 1, which indicates the first position in the reference weight value list, y(x) represents the x-th reference weight value in the reference weight value list, k represents a value configured based on experience, and is not limited here, for example, k may be half of the total number of reference weight values in the reference weight value list or another value, as long as k is less than t, where t is the total number of reference weight values in the reference weight value list, a1 represents the first weight conversion factor, a2 represents the second weight conversion factor, s1 represents the first initial position of the weight conversion, and s2 represents the second initial position of the weight conversion.

[000193] Clip3 is used to limit the reference weight values between a minimum and a maximum value, and the minimum and maximum values can be configured based on experience. For simplicity, the following description is given for a minimum value of 0 and a maximum value of 8 as an example.

[000194] a1 and a2 represent weight conversion factors that can be configured based on experience. For example, a1 can be an integer different from zero, for example, a1 can be -4, -3, -2, -1, 1, 2, 3 or 4, etc., a2 can be a non-zero integer, for example, a2 can be -4, -3, -2, -1, 1, 2, 3 or 4, etc. For example, when a1 is a positive integer, a2 can be a negative integer; when a1 is a negative integer, a2 can be a positive integer. For example, a1 can be -a2, that is, the rates of change are the same, which is reflected in the setting of the reference weight values, that is, the reference weight values have the same gradually changing width. For simplicity of description, take a1 as 1 and a2 as -1 as an example, the reference weight values should pass through 0, 1, 2, 3, 4, 5, 6, 7, 8 from 0 to 8, and then pass through 8, 7, 6, 5, 4, 3, 2, 1, 0 from 8 to 0. Optionally, the reference weight values pass through 8, 6, 4, 2, 0 from 8 to 0, and then pass through 0, 2, 4, 6, 8 from 0 to 8.

[000195] s1 and s2 may represent initial positions of the transformation of weights, which may be configured based on experience. For example, s1 may be an initial position of the transformation of reference weight values in the interval [0, k] and may be half of k; or s1 may be slightly less than half of k, for example, half of k minus 4; or s1 may be slightly greater than half of k, for example, half of k plus 4. The above examples are illustrative only, and the meaning of s1 is not limited here. s2 may be an initial position of the transformation of reference weight values in the interval [k+1, t], and s2 may be half of q (for example, the difference between t and k+1); or s2 may be slightly less than half of q, for example, half of q minus 4; or s2 may be slightly greater than half of q, for example, half of q plus 4. The above examples are illustrative only, and the meaning of s2 is not limited here.

[000196] In short, a list of reference weight values of the current block may be obtained, and the list of reference weight values may include a plurality of reference weight values that first monotonically increase and then monotonically decrease, or first monotonically decrease and then monotonically increase. In a possible implementation, the list of reference weight values may also include one or more reference weight values of first target positions, one or more reference weight values of second target positions, one or more reference weight values of first neighboring positions neighboring only the first target positions, one or more reference weight values of second neighboring positions neighboring both the first target positions and the second target positions, and one or more reference weight values of third neighboring positions neighboring only the second target positions.

[000197] The first target positions include one or more reference weight values determined based on the first initial position of the weight transformation. For example, based on the first initial position of the weight transformation, one reference weight value is determined, which is taken as the first target positions. Optionally, based on the first initial position of the weight transformation, a plurality of reference weight values are determined, which are taken as the first target positions. If the first target positions include a plurality of reference weight values, the plurality of reference weight values in the first target positions monotonically increase or decrease. The monotonous increase may be a monotonous increase of the plurality of reference weight values of the first target positions, and the monotonous decrease may be a monotonous decrease of the plurality of reference weight values of the first target positions.

[000198] The second target positions include one or more reference weight values determined based on the second initial position of the weight transformation. For example, based on the second initial position of the weight transformation, one reference weight value is determined, which is taken as the second target positions. Optionally, based on the second initial position of the weight transformation, a plurality of reference weight values are determined, which are taken as the second target positions. If the second target positions include a plurality of reference weight values, the plurality of reference weight values at the second target positions monotonically increase or decrease. The monotonous increase may be a strictly monotonous increase (the plurality of reference weight values of the second target positions strictly monotonically increase).

[000199] If the plurality of reference weight values of the first target positions monotonically increase (for example, strictly monotonically increase), the plurality of reference weight values of the second target positions monotonically decrease (for example, strictly monotonically decrease). Optionally, if the plurality of reference weight values of the first target positions monotonically decrease (for example, strictly monotonically decrease), the plurality of reference weight values of the second target positions monotonically increase (for example, strictly monotonically increase).

[000200] For example, all the reference weight values of the first adjacent positions are the first reference weight values, all the reference weight values of the second adjacent positions are the second reference weight values, and all the reference weight values of the third adjacent positions are the third reference weight values. The first reference weight values may be the same as the third reference weight values, the first reference weight values may differ from the second reference weight values, and the third reference weight values may differ from the second reference weight values. For example, all the reference weight values of the first adjacent positions are 0, all the reference weight values of the second adjacent positions are 8, and all the reference weight values of the third adjacent positions are 0. Optionally, all the reference weight values of the first adjacent positions are 8, all the reference weight values of the second adjacent positions are 0, and all the reference weight values of the third adjacent positions are 8.

[000201] Optionally, all the reference weight values of the first adjacent positions are the first reference weight values; the reference weight values of the second adjacent positions monotonically decrease; and the reference weight values of the third adjacent positions monotonically increase. For example, all the reference weight values of the first adjacent positions are 8, the first target positions include one reference weight value of 7, the reference weight values of the second adjacent positions monotonically decrease from 6 to 0, the second target positions include one reference weight value equal to 1, and the reference weight values of the third adjacent positions monotonically increase from 2 to 8.

[000202] Optionally, the reference weight values of the first adjacent positions monotonically decrease; the reference weight values of the second adjacent positions first monotonically decrease and then monotonically increase; the reference weight values of the third adjacent positions monotonically increase. For example, the reference weight values of the first adjacent positions monotonically decrease from 8 to 5, the first target positions include one reference weight value of 4, the reference weight values of the second adjacent positions first monotonically decrease from 3 to 0 and then monotonically increase from 0 to 3, the second target positions include one reference weight value of 4, and the reference weight values of the third adjacent positions monotonically increase from 5 to 8.

[000203] Optionally, the reference weight values of the first adjacent positions monotonically decrease; the reference weight values of the second adjacent positions monotonically increase; all the reference weight values of the third adjacent positions are third reference weight values. For example, the reference weight values of the first adjacent positions monotonically decrease from 8 to 1, the first target positions include one reference weight value of 0, the reference weight values of the second adjacent positions monotonically increase from 0 to 7, the second target positions include one reference weight value of 8, and all the reference weight values of the third adjacent positions are 8.

[000204] Optionally, all the reference weight values of the first adjacent positions are the first reference weight values; the reference weight values of the second adjacent positions monotonically increase; the reference weight values of the third adjacent positions monotonically decrease. For example, all the reference weight values of the first adjacent positions are 0, the first target positions include one reference weight value 1, the reference weight values of the second adjacent positions monotonically increase from 2 to 8, the second target positions include one reference weight value 7, and the reference weight values of the third adjacent positions monotonically decrease from 6 to 0.

[000205] Optionally, the reference weight values of the first adjacent positions monotonically increase; the reference weight values of the second adjacent positions first monotonically increase and then monotonically decrease; the reference weight values of the third adjacent positions monotonically decrease. For example, the reference weight values of the first adjacent positions monotonically increase from 0 to 3, the first target positions include one reference weight value of 4, the reference weight values of the second adjacent positions first monotonically increase from 5 to 8 and then monotonically decrease from 8 to 5, the second target positions include one reference weight value of 4, and the reference weight values of the third adjacent positions monotonically decrease from 3 to 0.

[000206] Optionally, the reference weight values of the first adjacent positions monotonically increase; the reference weight values of the second adjacent positions monotonically decrease; all the reference weight values of the third adjacent positions are third reference weight values. For example, the reference weight values of the first adjacent positions monotonically increase from 0 to 7, the first target positions include one reference weight value of 8, the reference weight values of the second adjacent positions monotonically decrease from 8 to 1, the second target positions include one reference weight value of 0, and all the reference weight values of the third adjacent positions are 0.

[000207] The above examples are illustrative only, and no limitations are made here, provided that the plurality of reference weight values in the reference weight value list can satisfy the following requirements: increasing from 0 to 8 and then decreasing from 8 to 0; or decreasing from 8 to 0 and then increasing from 0 to 8.

[000208] Case 3: The plurality of reference weight values in the reference weight value list may first include the plurality of first reference weight values and then include the plurality of second reference weight values, or may first include the plurality of second reference weight values and then include the plurality of first reference weight values. For example, the reference weight value list may be [8 8…8 8 0 0…0 0] or [0 0…0 0 8 8…8 8].

[000209] For example, the reference weight values in the reference weight value list may be pre-configured or configured based on weight configuration parameters. The weight configuration parameters may include a weight conversion start position. The weight conversion start position may be a value established based on experience.

[000210] In the process of pre-configuring the reference weight values in the reference weight value list, a plurality of reference weight values in the reference weight value list may be arbitrarily configured, provided that the plurality of reference weight values includes the first reference weight values and the second reference weight values.

[000211] In the process of configuring reference weight values in the reference weight value list based on the weight configuration parameters, a weight conversion start position may first be obtained, and then a plurality of reference weight values in the reference weight value list are determined based on the weight conversion start position. For example, the weight conversion start position represents the s-th reference weight value in the reference weight value list. Therefore, all reference weight values up to the s-th reference weight value (except for the s-th reference weight value) are the first reference weight values (for example, 8), and all reference weight values after the s-th reference weight value (including the s-th reference weight value) are the second reference weight values (for example, 0). Optionally, all reference weight values up to the s-th reference weight value (except the s-th reference weight value) are second reference weight values (e.g., 0), and all reference weight values after the s-th reference weight value (including the s-th reference weight value) are first reference weight values (e.g., 8).

[000212] Based on the above cases, a list of reference weight values of the current block can be obtained. For simplicity, in the following embodiments of the invention, the description is given for the list of reference weight values of case 1 as an example, and the lists of reference weight values of other cases can be implemented in a similar manner.

[000213] In step S2, the actual quantity is determined based on the size of the current block and the weighted prediction angle of the current block.

[000214] For example, the actual number refers to the presence of an actual number of surrounding positions outside the current block, for example, the pixel positions within the current block can only indicate an actual number of surrounding positions, that is, the target weight value of each of the pixel positions within the current block can be obtained by setting reference weight values for the actual number of surrounding positions.

[000215] For example, the number of surrounding positions beyond the current block is determined based on the size of the current block and/or the weighted prediction angle of the current block. In this embodiment of the invention, the number of surrounding positions beyond the current block is designated as the valid number ValidLength.

[000216] For example, the valid quantity can be defined by the following formula: ValidLength = (N+(M>>X))<<1 or ValidLength = (M+(N>>X))<<1. In the above formula, ValidLength represents the valid quantity, M represents the width of the current block, N represents the height of the current block, X represents the logarithm log2 value of the absolute value of the slope of the weighted prediction angle of the current block, such as 0 or 1, << refers to the arithmetic left shift, and >> refers to the arithmetic right shift.

[000217] In the present invention, a<<b can be understood as an arithmetic shift to the left of the integer binary-complement representation of a by b binary digits, and this operation is defined when b is a positive number. Simply put, a<<b can be understood as multiplying a by 2 to the power of b. In the present invention, a>>b can be understood as an arithmetic shift to the right of the integer binary-complement representation of a by b binary digits, and this operation is defined when b is a positive number. Simply put, a>>b can be understood as dividing a by 2 to the power of b.

[000218] In step S3, based on the size of the current block, the weighted prediction angle of the current block, and the weighted prediction position of the current block, a target index is determined.

[000219] For example, the target index may refer to the numbering of a reference weight value in a list of reference weight values, such that when the target index is 259, it may point to the 259th reference weight value in the list of reference weight values.

[000220] For example, the target index can be determined using the following formulas: FirstIndex=(HalfLength-4)-((ValidLength>>1)-a + Y * ((ValidLength - 1) >> 3)); or FirstIndex=(HalfLength-4)-((ValidLength>>1)-b+Y * ((ValidLength - 1)>>3) - ((M << 1) >> X)); or FirstIndex = (HalfLength - 4) - ((ValidLength>>1) - c + Y * ((ValidLength - 1) >> 3) - ((N << 1) >> X)); or FirstIndex = (HalfLength - 4) - ((ValidLength >> 1) - d+Y * ((ValidLength - 1) >> 3)). In the above formulas, FirstIndex refers to the target index, ValidLength refers to the valid quantity which is determined based on the size of the current block and the weighted prediction angle of the current block, Y refers to the position of the weighted prediction of the current block, the value of HalfLength - 4 refers to the starting position of the weight transformation, M refers to the width of the current block, N refers to the height of the current block, X is the logarithm of log2 of the absolute value of the slope of the weighted prediction angle of the current block, and a, b, c and d are preset constant values.

[000221] In the above process, the ValidLength value is related to the weighted prediction angle of the current block and the size of the current block. To simplify the scheme, some parameters may be fixed to achieve optimization, for example, the weighted prediction angle of the current block may be set to a fixed parameter value, and ValidLength is related only to the size of the current block. In other embodiments of the invention, ValidLength may be determined in a similar manner.

[000222] In the above process, the FirstIndex value is related to the weighted prediction angle of the current block, the size of the current block, and the weighted prediction position of the current block. In order to simplify the scheme, some parameters may be fixed to achieve optimization, for example, the weighted prediction angle of the current block may be set to a fixed parameter value, and the FirstIndex value is related only to the size of the current block and the weighted prediction position of the current block. Optionally, the weighted prediction angle of the current block and the weighted prediction position of the current block may be set to fixed parameter values, which may be the same or different for different current blocks, and the FirstIndex value is related only to the size of the current block. In other embodiments of the invention, FirstIndex (or FirstPos) can be determined in a similar manner, which will not be repeated here.

[000223] In step S4, an actual number of reference weight values is selected from the list of reference weight values based on the target index.

[000224] For example, it is assumed that the target index is q1 and the actual quantity is r. If the target index is the first reference weight value to be selected in the list of reference weight values, the reference weight values q1-th to q2-e are selected from the list of reference weight values, and the difference between the index related to q2 and the index related to q1 is r - 1, and thus r reference weight values are selected from the list of reference weight values. Alternatively, if the target index is the last reference weight value to be selected in the list of reference weight values, the reference weight values q3-th to q1-e are selected from the list of reference weight values, and the difference between the index related to q1 and the index related to q3 is r - 1, and thus r reference weight values are selected from the list of reference weight values.

[000225] The above methods are only examples, and the target index may also be the middle reference weight value to be selected in the list of reference weight values, the implementation is similar and will not be repeated here. The following description is given for the target index, which is the first reference weight value to be selected in the list of reference weight values, as an example, that is, the reference weight values q1-th to q2-e are selected.

[000226] In step S5, based on the actual number of reference weight values, reference weight values of surrounding positions outside the current block are set.

[000227] For example, the number of surrounding positions outside the current block is an actual number, and the actual number of reference weight values is selected from the reference weight value list. Thus, the number of surrounding positions is equal to the number of selected reference weight values, and the actual number of reference weight values in the reference weight value list can be set as reference weight values of the surrounding positions outside the current block, respectively.

[000228] For example, for a first reference weight value within the actual number of reference weight values, the first reference weight value is set as a reference weight value for a first surrounding position outside the current block; for a second reference weight value within the actual number of reference weight values, the second reference weight value is set as a reference weight value for a second surrounding position outside the current block, and so on.

[000229] In a possible implementation, if r reference weight values are selected from the list of reference weight values, r reference weight values are taken from the list of reference weight values and set as reference weight values of r surrounding positions outside the current block, respectively. Alternatively, if r reference weight values are selected from the list of reference weight values, instead of retrieving r reference weight values from the list of reference weight values, r reference weight values are used as reference weight values of r surrounding positions outside the current block by moving the reference weight values in the list of reference weight values.

[000230] Embodiment 7 of the invention. In the above Embodiments 1 to 3 of the invention, for each pixel position, based on a reference weight value associated with the surrounding corresponding position specified by the pixel position, a target weight value of the pixel position is determined. Below, as an example, a description is given for the surrounding corresponding position of the current block. The reference weight value associated with the surrounding corresponding position can be obtained as follows: a reference weight value is set directly for each of the surrounding positions outside the current block, for example, an actual number of reference weight values (not selected from the list of reference weight values) is obtained, and based on the actual number of reference weight values, reference weight values of the surrounding positions outside the current block are set.

[000231] For example, reference weight values configured for surrounding positions outside the current block may be pre-configured or configured based on weight configuration parameters. The weight configuration parameters may include a weight transformation coefficient and a weight transformation start position. For the weight transformation start position of each current block, the weight transformation start position is determined based on at least one of the following parameters: a weighted prediction angle of the current block, a weighted prediction position of the current block, or a size of the current block. When the current block includes one sub-block, the number of surrounding positions outside the current block (e.g., an actual number) is determined based on the size of the current block and the weighted prediction angle of the current block.

[000232] Thus, since reference weight values have been set for surrounding positions outside the current block, that is, each surrounding position has a reference weight value, after the surrounding corresponding position indicated by the pixel position is determined from the surrounding positions outside the current block, a reference weight value associated with the surrounding corresponding position, for example, a target weight value of the pixel position, can be determined.

[000233] The process of setting reference weight values of surrounding positions is described below in conjunction with implementations.

[000234] First, an actual number of reference weight values is obtained. Then, based on the actual number of reference weight values, reference weight values of the surrounding positions outside the current block are set, wherein the reference weight values of the surrounding positions outside the current block may monotonically increase or monotonically decrease.

[000235] It should be noted that the actual number of reference weight values is also obtained in the previous embodiment of the invention, and the actual number in the previous embodiment of the invention may be the same as or different from the actual number in this embodiment of the invention. For the purpose of distinction, the actual number in the previous embodiment of the invention may be considered as the first actual number, and the actual number in this embodiment of the invention may be considered as the second actual number.

[000236] For example, the number of surrounding positions beyond the current block is a valid number, and a valid number of reference weight values must be obtained. For example, the valid number can be determined by the following formula: ValidLength=(N+(M >> X)) << 1, where N and M respectively represent the height and width of the current block, and X represents the logarithm log2 value of the absolute value of the slope of the weighted prediction angle of the current block, such as 0 or 1.

[000237] In a possible implementation, the actual number of reference weight values may monotonically increase or monotonically decrease. Optionally, the actual number of reference weight values may first monotonically increase and then monotonically decrease, or first monotonically decrease and then monotonically increase. Optionally, the actual number of reference weight values may first include a plurality of first reference weight values and then include a plurality of second reference weight values, or may first include a plurality of second reference weight values and then include a plurality of first reference weight values. The description below will be given in combination with several cases.

[000238] Case 1: The actual number of reference weight values may increase monotonically or decrease monotonically.

[000239] For example, the reference weight values may be pre-configured or configured based on the weight configuration parameters. The weight configuration parameters may include a weight conversion factor and a weight conversion start position. The weight conversion factor may be a value set based on experience, and the weight conversion start position may also be a value set based on experience, or the weight conversion start position may also be determined based on the weighted prediction position or determined based on the weighted prediction angle and the weighted prediction position.

[000240] For example, in the process of pre-configuring a plurality of reference weight values, the plurality of reference weight values may be configured arbitrarily, provided that the plurality of reference weight values monotonically increase or monotonically decrease. Furthermore, in the process of configuring a plurality of reference weight values, based on the weight configuration parameters, a weight transformation coefficient and a weight transformation start position may first be obtained, and then a plurality of reference weight values are determined based on the weight transformation coefficient and the weight transformation start position. The weight transformation start position is determined based on the weighted prediction position of the current block, or is determined based on the weighted prediction angle and the weighted prediction position of the current block.

[000241] For example, the reference weight values may be defined as follows: y(x)=Clip3(minimum, maximum, a*(x-s)), where x represents an index of a surrounding position that is in the range from 1 to the value of the actual quantity, for example, x is 1, which indicates the first surrounding position, and y(x) represents the reference weight value of the x-th surrounding position, “a” represents a weight transformation factor, and s represents the starting position of the weight transformation.

[000242] The Clip3 function is used to limit the reference weight values between the minimum and maximum values. The maximum value and the minimum value can be configured based on experience. For simplicity, the following description is given for the minimum value of 0 and the maximum value of 8 as an example.

[000243] "a" represents a weight transformation coefficient that can be configured based on experience, for example, "a" can be a non-zero integer, for example, "a" can be -4, -3, -2, -1, 1, 2, 3 or 4, etc. For simplicity, the description is given for "a" equal to 1 as an example. If the value of "a" is 1, the reference weight values should pass through 0, 1, 2, 3, 4, 5, 6, 7, 8 from 0 to 8 or pass through 8, 7, 6, 5, 4, 3, 2, 1, 0 from 8 to 0. For example, when "a" is a positive integer, "a" can be positively correlated with the number of surrounding positions, that is, the more surrounding positions outside the current block, the greater the value of "a". When "a" is a negative integer, "a" may have a negative correlation with the number of surrounding positions, i.e. the more surrounding positions there are outside the current block, the smaller the value of "a". The above are examples of the value of "a", and no restrictions are made here.

[000244] s may represent a starting position of the weight transformation, which may be determined based on the weighted prediction position. For example, s=f (weighted prediction position), that is, s is a function of the weighted prediction position. For example, after determining the range of surrounding positions outside the current block, the actual number of surrounding positions is determined, and all surrounding positions are equally divided into N parts, where the value of N can be arbitrarily set, such as 4, 6, or 8, etc. The weighted prediction position is used to represent which surrounding position outside the current block is taken as the target surrounding positions of the current block, and the surrounding position corresponding to the weighted prediction position is the starting position of the weight transformation. Optionally, the value of s may be determined based on the weighted prediction angle and the weighted prediction position, for example, s=f (weighted prediction angle and weighted prediction position), that is, s is a function of the weighted prediction angle and the weighted prediction position. For example, the range of surrounding positions outside the current block can be determined based on the weighted prediction angle. After the range of surrounding positions outside the current block is determined, the actual number of surrounding positions can be determined, and all surrounding positions are equally divided into N parts. The weighted prediction position is used to represent which surrounding position outside the current block is taken as the target surrounding positions of the current block, and the surrounding position corresponding to the weighted prediction position is the starting position of the weight transformation.

[000245] To summarize, in y(x)=Clip3(minimum, maximum, a*(x-s)) the weight transformation coefficient a and the initial position s of the weight transformation are known values. For each of the surrounding positions outside the current block, a reference weight value of the surrounding position can be determined based on a functional relationship. For example, if the weight transformation coefficient a is 2 and the initial position s of the weight transformation is 2, the functional relationship is y(x)=Clip3(minimum, maximum, 2*(x-2)). For each surrounding position x outside the current block, a reference weight value y can be obtained.

[000246] In this way, it is possible to obtain an actual number of reference weight values of the current block, and these reference weight values may monotonically increase or monotonically decrease. In a possible implementation, the reference weight values of the surrounding positions outside the current block include one or more reference weight values of the target positions, one or more reference weight values of the first neighboring positions of the target positions, and one or more reference weight values of the second neighboring positions of the target positions.

[000247] The target positions, the reference weight values of the target positions, the reference weight values of the first adjacent positions, and the reference weight values of the second adjacent positions may be referred to in the description of case 1 in the previous embodiment of the invention and will not be repeated here.

[000248] Case 2: The actual number of reference weight values first increases monotonically and then decreases monotonically, or first decreases monotonically and then increases monotonically.

[000249] For example, the actual number of reference weight values may be pre-configured or configured based on weight configuration parameters. The weight configuration parameters may include a weight conversion factor and a weight conversion start position. The weight conversion factor may be a value set based on experience, and the weight conversion start position may also be a value set based on experience, or determined based on a weighted prediction position, or determined based on a weighted prediction angle and a weighted prediction position.

[000250] For example, in the process of pre-configuring a plurality of reference weight values, the plurality of reference weight values may be arbitrarily configured on the condition that the plurality of reference weight values first monotonically increase and then monotonically decrease, or first monotonically decrease and then monotonically increase. In the process of configuring a plurality of reference weight values, based on the weight configuration parameters, a first weight conversion factor, a second weight conversion factor, a first weight conversion start position, and a second weight conversion start position may first be obtained, and then a plurality of reference weight values are determined based on the first weight conversion factor, the second weight conversion factor, the first weight conversion start position, and the second weight conversion start position.

[000251] For example, a plurality of reference weight values may be defined as follows: when x is in [0, k], y(x)=Clip3(minimum, maximum, a1*(x-s1)). When x is in [k+1, t], y(x)=Clip3(minimum, maximum, a2*(x-s2)). X represents a position index of a surrounding position, for example, the value of x is 1, which indicates the first surrounding position, and y(x) represents a reference weight value of the x-th surrounding position. K is a value configured based on experience, which is not limited here, for example, k may be half of the actual quantity or another value if k is less than t, where t is the total number of surrounding positions, that is, the above-mentioned actual quantity, a1 represents the first weight transformation coefficient, a2 represents the second weight transformation coefficient, s1 represents the first starting position of the weight transformation, and s2 represents the second starting position of the weight transformation.

[000252] Clip3 is used to limit the reference weight values between a minimum and a maximum value, and the minimum and maximum values can be configured based on experience. For simplicity, the following description is given for a minimum value of 0 and a maximum value of 8 as an example.

[000253] a1 and a2 represent weight conversion factors that can be configured based on experience. For example, a1 may be an integer different from zero, such as al may be -4, -3, -2, -1, 1, 2, 3, or 4, etc. a2 may be a non-zero integer, such as a2 may be -4, -3, -2, -1, 1, 2, 3, or 4, etc.

[000254] s1 and s2 may represent initial positions of weight transformation that may be configured based on experience, s1 may be an initial position of transformation of reference weight values in the interval [0, k], and s2 may be an initial position of transformation of reference weight values in the interval [k+1, t].

[000255] s1 may be determined based on the weighted prediction position. For example, s1=f(weighted prediction position), that is, s1 is a function of the weighted prediction position. For example, after determining the range of surrounding positions outside the current block, the range [0, k] is determined from all surrounding positions, and all surrounding positions in the range [0, k] are equally divided into N parts, where the value of N can be set arbitrarily. The weighted prediction position is used to represent which surrounding position within the range [0, k] is taken as the target surrounding positions of the current block, and the surrounding position corresponding to the weighted prediction position is the starting position s1 of the weight transformation. Optionally, s1 may be determined based on the weighted prediction angle and the weighted prediction position, for example, s1=f(weighted prediction angle, weighted prediction position), that is, s1 is a function of the weighted prediction angle and the weighted prediction position. For example, the range of surrounding positions beyond the current block can be determined based on the weighted prediction angle, and the range [0, k] is determined from all surrounding positions. All surrounding positions in the range [0, k] are equally divided into N parts. The weighted prediction position is used to represent which surrounding position in the range [0, k] is taken as the target surrounding positions of the current block to obtain the initial position s1 of the weight transformation.

[000256] s2 may be determined based on the weighted prediction position, or may be determined based on the weighted prediction angle and the weighted prediction position. The process of determining s2 may be related to the process of determining s1, except that the range is changed to [k+1, t], and therefore is not repeated here.

[000257] Examples of determining the initial positions s1 and s2 of the weight transformation are given above, and no restrictions are made here.

[000258] In this manner, a plurality of reference weight values may be obtained, and these reference weight values may first monotonically increase and then monotonically decrease, or first monotonically decrease and then monotonically increase. In a possible implementation, the reference weight values of a plurality of surrounding positions outside the current block may also include one or more reference weight values of first target positions, one or more reference weight values of second target positions, one or more reference weight values of first neighboring positions neighboring only the first target positions, one or more reference weight values of second neighboring positions neighboring both the first target positions and the second target positions, and one or more reference weight values of third neighboring positions neighboring only the second target positions.

[000259] For the first target positions, the second target positions, the reference weight values of the first target positions, the reference weight values of the second target positions, the reference weight values of the first adjacent positions, the reference weight values of the second adjacent positions, and the reference weight values of the third adjacent positions, etc., reference may be made to the description of Case 2 in the previous embodiment of the invention, and redundant descriptions are not given here.

[000260] Case 3: The actual number of reference weight values may first include a plurality of first reference weight values and then a plurality of second reference weight values, or first include a plurality of second reference weight values and then a plurality of first reference weight values. The actual number of reference weight values may be pre-configured or configured based on the weight configuration parameters. The weight configuration parameters may include a weight conversion start position. In the process of configuring the reference weight values based on the weight configuration parameters, a weight conversion start position may be obtained, and the plurality of reference weight values are determined based on the weight conversion start position.

[000261] Based on several cases described above, an actual number of reference weight values can be obtained. For simplicity, in the following embodiments of the invention, the description is given for the reference weight value in case 1 as an example, and the reference weight values in other cases are implemented in a similar manner.

[000262] After the actual number of reference weight values is obtained, the reference weight values of the surrounding positions outside the current block are set based on the actual number of reference weight values, and the reference weight values of the surrounding positions outside the current block may monotonically increase or monotonically decrease.

[000263] For example, the number of surrounding positions outside the current block is an actual number, and the number of reference weight values is an actual number. Therefore, the actual number of reference weight values can be set as reference weight values of surrounding positions outside the current block.

[000264] For example, the first reference weight value is set as the reference weight value of the first surrounding position outside the current block, the second reference weight value is set as the reference weight value of the second surrounding position outside the current block, etc.

[000265] Embodiment 8 of the invention. In the above embodiments 1 to 7 of the invention, the current block may include one sub-block, that is, the sub-block is the current block itself. For the current block, M and N represent the width and height of the current block (that is, the current prediction unit), respectively. In this case, as shown in Fig. 6B, the method for obtaining a weight array for angle weighted prediction (AWP) may include steps a1 to a4.

[000266] In step a1, parameters such as stepIdx, angleIdx, and subAngleIdx are obtained based on AwpIdx.

[000267] AwpIdx represents the value of the weighted prediction angle index and the weighted prediction position. If there are 7 weighted prediction positions and 8 weighted prediction angles, the range of AwpIdx values is from 0 to 55. If the weighted prediction position is from -3 to 3 (representing the fourth weighted prediction position is the center, and the fourth weighted prediction position is 0), and if the weighted prediction angle index is 0-7, the weighted prediction positions and weighted prediction angles corresponding to the 56 values of the AwpIdx index are shown in Table 2.

[000268] stepIdx represents a weighted prediction position that is in the range from -3 to 3. For example, for the first weighted prediction position, the weighted prediction position is -3; for the second weighted prediction position, the weighted prediction position is -2, and so on; for the seventh weighted prediction position, the weighted prediction position is 3.

[000269] angleIdx is a value of the logarithm log2 of the absolute value of the slope of the weighted prediction angle (e.g., 0, or 1, or a larger constant), and subangleIdx is an angular region in which the weighted prediction angle is located. In Fig. 5B, eight weighted prediction angles are shown, where angleIdx of weighted prediction angle 0 is a value of the logarithm log2 of the absolute value of the slope of weighted prediction angle 0, angleIdx of weighted prediction angle 1 is a value of the logarithm log2 of the absolute value of the slope of weighted prediction angle 1, etc., and angleIdx of weighted prediction angle 7 is a value of the logarithm log2 of the absolute value of the slope of weighted prediction angle 7. Weighted prediction angle 0 and weighted prediction angle 1 are located in angular region 0, weighted prediction angle 2 and weighted prediction angle 3 are located in angular region 1, weighted prediction angle 4 and weighted prediction angle 5 are located in angular region 2, and weighted prediction angle 6 and weighted prediction angle 7 are located in angular region 3.

[000270] For example, stepIdx can be determined using the following formula: stepIdx=(AwpIdx>>3)-3.

[000271] For example, modAngNum(angle index) may first be determined based on the following formula: modAngNum=AwpIdx%8; then angleIdx is determined based on modAngNum as follows: if the value of modAngNum is 2, angleIdx = 7; if the value of modAngNum is 6, angleIdx = 8; otherwise, angleIdx = modAngNum%2.

[000272] For example, subangleIdx can be determined using the following formula: subangleIdx=modAngNum>>1.

[000273] Thus, after determining the weighted prediction angle of the current block and the weighted prediction position of the current block, the encoder side can determine the AwpIdx value based on the weighted prediction angle and the weighted prediction position, as shown in Table 2. When the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may contain the AwpIdx value. Thus, the decoder side can obtain the AwpIdx value, and then obtain the stepIdx, angleIdx, and subangleIdx based on the AwpIdx.

[000274] For example, based on angleIdx and subangleIdx, one weighted prediction angle can be uniquely determined, as shown in Table 3. The weighted prediction angle can also be determined in another way, such as by changing the section number, etc.

[000275] In step a2, based on stepIdx, angleIdx and subangleIdx, a list of reference weight values ReferenceWeight[x] is configured, which means setting reference weight values for surrounding positions outside the current block. Based on subangleIdx, ReferenceWeight[x] can be calculated in the following several cases.

[000276] Case 1: If the value of subangleIdx is 0, that is, the weighted prediction angle is in the angle region of 0, for example, the weighted prediction angle is equal to weighted prediction angle 0 or weighted prediction angle 1, the initial position FirstPos of the weight transformation can be determined by the following formula: FirstPos = (ValidLength_H>>1)-6+DeltaPos_H. Then, the reference weights of the surrounding positions outside the current block can be determined by the following formula: ReferenceWeights[x]=Clip3(0, 8, x-FirstPos). The formula is given for the minimum value of reference weights being 0, the maximum value of reference weights being 8, and the weight conversion factor being 1 as an example, that is, the above formula can be equivalent to ReferenceWeights[x]=Clip3(minimum, maximum, a*(x-FirstPos)), where x can refer to the index of surrounding positions beyond the current block, x ranges from 0 to ValidLength_H-1, and “a” represents the weight conversion factor.

[000277] In the above formula, ValidLength_H represents the number of surrounding positions beyond the current block (i.e., the valid number, also called the valid length). When the subangleIdx value is 0, the surrounding positions beyond the current block specified by the weighted prediction angle may be the surrounding positions in the left column. Therefore, the valid number is denoted as ValidLength_H. For example, the valid number ValidLength_H can be determined by the following formula: ValidLength_H=(N+(M>>angleIdx)) << 1, the left shift by one bit is due to the fact that the formula uses 1/2 pixel precision. If 1 pixel precision is used, the formula is: ValidLength_H=(N+(M >> angleIdx)). If 1/4 pixel precision is adopted, the formula is: ValidLength_H=(N+(M >> angleIdx)) << 2. If 2 pixel precision is adopted, ValidLength_H=(N+(M >> angleIdx)) >> 1. Other pixel precisions may be used, and no redundant descriptions are made here. In subsequent formulas, the operations involving >> 1 may change for different pixel precisions.

[000278] In the above formula, DeltaPos_H represents the position change parameter (for example, the intermediate parameter). When the subangleIdx value is 0, the surrounding positions beyond the current block specified by the weighted prediction angle may be the surrounding positions in the left column. Therefore, the position change parameter is denoted as DeltaPos_H. For example, DeltaPos_H is determined by the following formula: DeltaPos_H = stepIdx * ((ValidLength_H >> 3)-1).

[000279] Case 2: If the value of subangleIdx is 1, that is, the weighted prediction angle is in the angle region 1, for example, the weighted prediction angle is weighted prediction angle 2 or weighted prediction angle 3, the initial position FirstPos of the weight transformation can be determined by the following formula: FirstPos=(ValidLength_H >> 1) - 4 + DeltaPos_H - ((M << 1) >> angleIdx). Then, the reference weights of the surrounding positions outside the current block can be determined by the following formula: ReferenceWeights[x]=Clip3(0, 8, x-FirstPos). The formula is given for the minimum reference weight value of 0, the maximum reference weight value of 8, and the weight conversion factor of 1, as an example, x represents the index of the surrounding positions outside the current block, and x ranges from 0 to ValidLength_H-1.

[000280] In the formula above, ValidLength_H and DeltaPos_H can be assigned to case 1 and will not be repeated here.

[000281] Case 3: If the value of subangleIdx is 2, that is, the weighted prediction angle is in the angle region of 2, for example, the weighted prediction angle is weighted prediction angle 4 or weighted prediction angle 5, the initial position FirstPos of the weight transformation can be determined by the following formula: FirstPos=(ValidLength_W >> 1) - 4 + DeltaPos_W - ((N << 1) >> angleIdx). Then, the reference weights of the surrounding positions outside the current block can be determined by the following formula: ReferenceWeights[x]=Clip3(0, 8, x-FirstPos). The formula is given for the minimum reference weight value of 0, the maximum reference weight value of 8, and the weight conversion factor of 1, as an example, x represents the index of the surrounding positions beyond the current block, and x ranges from 0 to ValidLength_W - 1.

[000282] In the above formula, ValidLength_W represents the number of surrounding positions beyond the current block (i.e., the valid number, also called the valid length). When the subangleIdx value is 2, the surrounding positions beyond the current block specified by the weighted prediction angle can be the surrounding positions in one upper row. Therefore, the valid number is denoted as ValidLength_W. For example, the valid number ValidLength_W can be determined by the following formula: ValidLength_W=(M+(N >> angleIdx)) << 1.

[000283] In the above formula, DeltaPos_W represents the position change parameter (for example, the intermediate parameter). When the subangleIdx value is 2, the surrounding positions beyond the current block specified by the weighted prediction angle may be the surrounding positions in the upper row. Therefore, the position change parameter is denoted as DeltaPos_W. For example, DeltaPos_W is determined by the following formula: DeltaPos_W = stepIdx * ((ValidLength_W>> 3)-1).

[000284] Case 4: If the value of subangleIdx is 3, that is, the weighted prediction angle is in the angle region of 3, for example, the weighted prediction angle is the weighted prediction angle 6 or the weighted prediction angle 7, the initial position FirstPos of the weight transformation can be determined by the following formula: FirstPos=(ValidLength_W>>1)-6 + DeltaPos_W. Then, the reference weights of the surrounding positions outside the current block can be determined by the following formula: ReferenceWeights[x]=Clip3(0, 8, x-FirstPos). The formula is given for the minimum reference weight value of 0, the maximum reference weight value of 8, and the weight transformation coefficient of 1, as an example, x represents the index of the surrounding positions outside the current block and is in the range from 0 to ValidLength_W-1.

[000285] In the formula above, ValidLength_W and DeltaPos_W can be assigned to case 3 and will not be repeated here.

[000286] In this way, it is possible to determine which case to use based on subangleIdx, for example, in cases 1 and 2, ValidLength_H and DeltaPos_H are determined based on angleIdx and stepIdx, and FirstPos is determined based on ValidLength_H and DeltaPos_H, and then the reference weight values are set based on FirstPos. In cases 3 and 4, ValidLength_W and DeltaPos_W are determined based on angleIdx and stepIdx, and FirstPos is determined based on ValidLength_W and DeltaPos_W, and then the reference weight values are set based on FirstPos.

[000287] The formulas of the above cases differ as follows: when the upper left corner of the current block is used as the origin, the starting position of the list of reference weight values ReferenceWeights[x] is changed. For example, Fig. 6C shows examples of corner region 2 and corner region 3. When a precision of 1/2 pixel is adopted, the starting position of the list of reference weight values ReferenceWeights[x] is equal to (height<<1)>>angleIdx, that is, the offset in the formula is equal to "-((N<<1) >> angularldx)". Corner region 0 and corner region 1 are implemented in a similar manner, except that the offset in the formula is equal to "- ((M << 1) >> angleIdx)", that is, the height is changed to width.

[000288] In step a3, the pixel position brightness weight value is obtained based on angleIdx and the list of reference weight values ReferenceWeight[x]. Based on subangleIdx, the pixel position brightness weight value can be determined in the following several cases.

[000289] Case 1: If the value of subangleIdx is 0, the luminance weight value of the pixel position (x, y) can be determined by the following formula: AwpWeightArrayY[x][y]=Reference Weights[(y << 1) + ((x << 1) >> angleIdx)], where (y << 1) + ((x << 1) >> angleIdx) represents the surrounding position specified by the pixel position (x, y), ReferenceWeights[(y << 1) + ((x << 1) >> angleIdx)] represents the reference weight value of the surrounding position. The range of x is from 0 to M-1, and the range of y is from 0 to N-1.

[000290] Case 2: If the subangleIdx value is 1, the luminance weight value of the pixel position (x, y) can be determined by the following formula: AwpWeightArrayY[x][y] = Reference Weights [(y << 1) - ((x << 1) >> angleIdx)], where (y << 1) - ((x << 1) >> angleIdx) represents the surrounding position specified by the pixel position (x, y), Reference Weights [(y << 1) - ((x << 1) >> angleIdx)] represents the reference weight value of the surrounding position. The range of x values is from 0 to M-1, and the range of y values is from 0 to N-1.

[000291] Case 3: If the value of subangleIdx is 2, the brightness weight value of the pixel position (x, y) can be determined by the following formula: AwpWeightArrayY[x][y] = ReferenceWeights[(x << 1) - ((y << 1) >> angleIdx)], where (x << 1) - ((y << 1) >> angleIdx) represents the surrounding position specified by the pixel position (x, y), ReferenceWeights[(x << 1) - ((y << 1) >> angleIdx)] represents the reference weight value of the surrounding position. The range of x is from 0 to M-1, and the range of y is from 0 to N-1.

[000292] Case 4: If the value of subangleIdx is 3, the luminance weight value of the pixel position (x, y) can be determined by the following formula: AwpWeightArrayY[x][y]=ReferenceWeights[(x << 1)+((y << 1) >> angleIdx)], where (x << 1)+((y << 1) >> angleIdx) represents the surrounding position specified by the pixel position (x, y), ReferenceWeights[(x << 1)+((y << 1) >> angleIdx)] represents the reference weight value of the surrounding position. The range of x is from 0 to M-1, and the range of y is from 0 to N-1.

[000293] The above-mentioned steps a2 and a3 may be combined into one step, that is, step a2 (the list of reference weight values ReferenceWeight[x] is configured based on the stepIdx, angleIdx and subangleIdx, and configuring the list of reference weights ReferenceWeight[x] means setting reference weight values for the surrounding positions outside the current block) and step a3 (the weight value of the luminance of the pixel position is obtained based on the angleIdx and the list of reference weight values ReferenceWeight[x]) may be combined into a step of obtaining the weight value of the luminance of the pixel position based on the stepIdx, angleIdx and subangleIdx, that is, determining the weight value of the luminance of the pixel position based on the value of the coordinate of the surrounding corresponding position of the current block and the value of the coordinate of the starting position of the weight transformation of the current block.

[000294] For example, in case 1, if the value of subangleIdx is 0, that is, the weighted prediction angle is in the angular region of 0, for example, the weighted prediction angle is weighted prediction angle 0 or weighted prediction angle 1, the initial position FirstPos of the weight transformation can be determined by the following formula: FirstPos = (ValidLength_H >> 1) - 6 + DeltaPos_H. Then, the weight value of the luminance of the pixel position (x, y) is determined by the following formula: AwpWeightArrayY[x][y] = Clip3(0, 8, (y << 1) + ((x << 1) >> angleIdx)-FirstPos), where (y << 1)+((x << 1) >> angleIdx) represents the surrounding corresponding position specified by the pixel position (x, y). Other cases are similar.

[000295] In step a4, a chromaticity weight value for a pixel position is obtained based on a luminance weight value for the pixel position, and the luminance weight value for the pixel position and the chromaticity weight value for the pixel position may form a target pixel position weight value.

[000296] For example, if the 4:2:0 color resolution format is used, the chroma weight value of the pixel position (x, y) is determined by the following formula: AwpWeightArrayUV[x][y]=AwpWeightArrayY[x << 1][y << 1]. In another example, if the 4:4:4 color resolution format is used, the chroma weight value of the pixel position (x, y) is determined by the following formula: AwpWeightArrayUV[x][y] = AwpWeightArrayY[x] [y]. The range of x values is from 0 to M/2-1, and the range of y values is from 0 to N/2-1.

[000297] Step a4 may be implemented in another way: instead of obtaining the chroma weight value based on the luminance weight value, the chroma weight value for the pixel position is obtained based on the angleIdx and the list of reference weight values ReferenceWeight[x]. For example, if the 4:2:0 chroma resolution format is used, the chroma weight value for the pixel position may be obtained based on the angleIdx and the list of reference weight values ReferenceWeight[x].

[000298] For example, if the subangleIdx value is 0, the chromaticity weight value of the pixel position (x, y) is determined by the following formula: AwpWeightArrayUV[x][y] = Reference Weights [(y << 2)+((x << 2) >> angleIdx)].

[000299] For example, if the subangleIdx value is 1, the chromaticity weight value of the pixel position (x, y) is determined by the following formula: AwpWeightArrayUV[x][y] = Reference Weights[(y << 2) - ((x << 2) >> angleIdx)].

[000300] For example, if the subangleIdx value is 2, the chromaticity weight value of the pixel position (x, y) is determined by the following formula: AwpWeightArrayUV[x][y] = ReferenceWeights[(x << 2) - ((y << 2) >> angleIdx)].

[000301] For example, if the subangleIdx value is 3, the chromaticity weight value of the pixel position (x, y) is determined by the following formula: AwpWeightArrayUV[x][y] = Reference Weights[(x << 2) + ((y << 2) >> angleIdx)].

[000302] In the above formulas, the range of x values is from 0 to M-1, and the range of y values is from 0 to N-1.

[000303] In steps a3 and a4, the formulas in these cases differ as follows. Fig. 6D shows examples of corner region 2 and corner region 3. When the upper left corner of the current block is taken as the origin, the formula for calculating the surrounding corresponding position (x, y) in corner region 2 can be x - y >> angleIdx, and the formula for calculating the surrounding corresponding position (x, y) in corner region 3 can be x + y >> angleIdx. When a precision of 1/2 pixel is adopted, the formula for calculating the surrounding corresponding position (x, y) in corner region 2 can be (x << 1) - ((y << 1) >> angleIdx), and the formula for calculating the surrounding corresponding position (x, y) in corner region 3 can be (x << 1) + ((y << 1) >> angleIdx). Corner region 0 and corner region 1 can be implemented in a similar way, except that the x, y positions are swapped.

[000304] Embodiment 9 of the invention. In Embodiments 1 to 3 of the invention described above, the first pixel position prediction value is determined based on the first prediction mode, and the second pixel position prediction value is determined based on the second prediction mode.

[000305] In this embodiment, the description is given for the first prediction mode being the inter prediction mode and the second prediction mode being the inter prediction mode, as an example.

[000306] The first prediction mode is an inter prediction mode, the second prediction mode is an inter prediction mode, and a motion compensation candidate list is created, wherein the motion compensation candidate list includes at least two possible motion information. One possible motion information is selected from the motion compensation candidate list as the first target motion information of the current block, and the other possible motion information is selected from the motion compensation candidate list as the second target motion information of the current block. For each pixel position of the current block, a first pixel position prediction value is determined based on the first target motion information, and a second pixel position prediction value is determined based on the second target motion information.

[000307] For example, both the encoder side and the decoder side may create a motion compensation candidate list, and the motion compensation candidate list on the encoder side may be the same as the motion compensation candidate list on the decoder side, which is not limited here.

[000308] For example, all the possible motion information in the motion compensation candidate list is motion information with one hypothesis, for example, the possible motion information in the motion compensation candidate list is unidirectional motion information, not bidirectional motion information. Since all the possible motion information is motion information with one hypothesis, the motion compensation candidate list may be a unidirectional motion compensation candidate list.

[000309] For example, when a list of candidates for motion compensation is created, spatial motion information (e.g., one or more spatial motion vectors) may be added first, and then temporal motion information (e.g., one or more temporal motion vectors) may be added; and/or, when a list of candidates for motion compensation is created, unidirectional motion information (e.g., one or more unidirectional motion vectors) may be added first, and then bidirectional motion information (e.g., one or more bidirectional motion vectors) may be added, which is not limited here.

[000310] For example, when bidirectional motion information is added to the motion compensation candidate list, the bidirectional motion information is first divided into two unidirectional motion information, and the two divided unidirectional motion information is added to the motion compensation candidate list in order. Or, when bidirectional motion information is added to the motion compensation candidate list, the bidirectional motion information is first reduced to one unidirectional motion information, and then this one unidirectional motion information is added to the motion compensation candidate list. For example, reducing the bidirectional motion information to one unidirectional motion information includes: directly obtaining the unidirectional motion information from List0 (reference picture list 0); or directly obtaining the unidirectional motion information from List1 (reference picture list 1); or determining to obtain the unidirectional motion information from List0 or List1 based on the adding order.

[000311] The above are only examples of the motion compensation candidate list, and the motion information in the motion compensation candidate list is not limited here.

[000312] The encoder side may select, based on the RDO, one possible motion information from the motion compensation candidate list as the first target motion information of the current block and select another possible motion information from the motion compensation candidate list as the second target motion information of the current block, wherein the first target motion information is different from the second target motion information, which is not limited here.

[000313] In a possible implementation, when the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may carry indication information a and indication information b. The indication information a indicates an index value 1 of the first target motion information of the current block, and the index value 1 represents which possible motion information in the motion compensation candidate list is the first target motion information. The indication information b indicates an index value 2 of the second target motion information of the current block, and the index value 2 represents which possible motion information in the motion compensation candidate list is the second target motion information. For example, the index value 1 and the index value 2 may be different.

[000314] After receiving the coded bitstream, the decoder side may analyze the indication information a and the indication information b from the coded bitstream. Based on the indication information a, the decoder side selects the possible motion information corresponding to the index value 1 from the motion compensation candidate list as the first target motion information of the current block. Based on the indication information b, the decoder side selects the possible motion information corresponding to the index value 2 from the motion compensation candidate list as the second target motion information of the current block.

[000315] In another possible implementation, when the encoder side transmits the coded bitstream to the decoder side, the coded bitstream may carry indication information a and indication information c. The indication information a indicates an index value 1 of the first target motion information of the current block, and the index value 1 represents which possible motion information in the motion compensation candidate list is the first target motion information. The indication information c indicates a difference between the index value 1 and the index value 2, where the index value 2 represents which possible motion information in the motion compensation candidate list is the second target motion information. For example, the index value 1 and the index value 2 may be different.

[000316] After receiving the coded bitstream, the decoder side may analyze the indication information a and the indication information c from the coded bitstream. Based on the indication information a, the decoder side selects the possible motion information corresponding to index value 1 from the candidate list for motion compensation as the first target motion information of the current block. Based on the indication information c, the decoder side first determines index value 2 based on index value 1 and the difference between index value 1 and index value 2, and then selects the possible motion information corresponding to index value 2 from the candidate list for motion compensation as the second target motion information of the current block.

[000317] The process in which the encoder side/decoder side determines a first pixel position prediction value based on the first target motion information and determines a second pixel position prediction value based on the second target motion information may refer to an inter prediction process and will not be repeated here.

[000318] For example, when the first pixel position prediction value is determined based on the first target motion information, the first pixel position prediction value can be obtained using an inter weighted prediction mode. For example, an initial pixel position prediction value is determined based on the first target motion information, and then the initial prediction value is multiplied by a predetermined coefficient to obtain a prediction correction value. If the prediction correction value is greater than the maximum prediction value, the maximum prediction value is taken as the first prediction value of the current block; if the prediction correction value is less than the minimum prediction value, the minimum prediction value is taken as the first prediction value of the current block; and if the prediction correction value is not less than the minimum prediction value or is not greater than the maximum prediction value, the prediction correction value is taken as the first prediction value of the current block. The above-described method is only illustrative and is not limited here.

[000319] Similarly, when the second pixel position prediction value is determined based on the second target motion information, the second pixel position prediction value may be obtained using an inter weighted prediction mode. The specific implementation may be referred to the above example and will not be repeated here.

[000320] In the above cases, the indication information of the prediction information of the first prediction mode and the indication information of the prediction information of the second prediction mode may be interchangeable, provided that the encoder side and the decoder side are consistent. The exchange of the indication information does not affect the analysis process, that is, there is no dependence on the analysis. In the case of the same list of prediction mode candidates, the indication information of the prediction information of the first prediction mode cannot coincide with the indication information of the prediction information of the second prediction mode. It is assumed that two indices are encoded, where the index value a is 1 and the index value b is 3, when the index value a is encoded first, the index value b can be encoded as 2(3-1); when the index value b is encoded first, the index value b should be encoded as 3. In a word, the indication information of encoding with a smaller index value first can reduce the encoding overhead for encoding a larger index value. According to the method of creating the prediction mode candidate list, the first prediction mode is likely to come from the left side, and based on this previous experience, adjustment can be made on the encoder side and the decoder side, and the indication information of the prediction information in the area adjacent to the left side can be first encoded.

[000321] In the above case, the candidate list of the prediction mode may be a candidate list for motion compensation, the prediction information of the first prediction mode may be the first target motion information, and the prediction information of the second prediction mode may be the second target motion information. In the coded bitstream, the indication information of the first target motion information, such as the index value a, is first coded, and then the indication information of the second target motion information, such as the index value b, is coded. Alternatively, the indication information of the second target motion information, such as the index value b, is first coded, and then the indication information of the first target motion information, such as the index value a, is coded. For example, if the index value a is 1 and the index value b is 3, the index value a is coded before the index value b. In another example, if the index value b is 1 and the index value a is 3, the index value b is coded before the index value a.

[000322] Embodiment 10 of the invention. For embodiment 9 of the invention, a motion compensation candidate list should be created, which may be a unidirectional motion compensation candidate list. Fig. 7 is a schematic diagram of the current block and neighboring blocks. In the process of constructing a motion compensation candidate list based on the order of F, G, C, A, B, and D, instances of unidirectional motion information are added to the motion compensation candidate list in the priority order of List0, and a check for duplicate entries should be performed in the adding process. The order of F, G, C, A, B, and D is only an example, and another order may also be adopted, which is not limited in this description.

[000323] If the motion compensation candidate list is not complete, the remaining bidirectional motion information is further split into unidirectional motion information and added to the motion compensation candidate list in the order of List0 as a priority, and a check for duplicate entries must be performed during the adding process.

[000324] Finally, if the motion compensation candidate list is not complete, the temporary motion information is further split into unidirectional motion information and added to the motion compensation candidate list in the order of List0 as a priority, and a check for duplicate entries must be performed in the adding process.

[000325] The List0 strategy as a priority is as follows: if there is motion information of List0, the motion information of List0 is added to the motion compensation candidate list, and a check for duplicate entries must be performed during the adding process. If there is motion information of List1, the motion information of List1 is added to the motion compensation candidate list, and a check for duplicate entries must be performed during the adding process.

[000326] The process of creating a list of candidates for motion compensation is described below in conjunction with several specific application scenarios.

[000327] Application scenario 1: in step 1, as shown in Fig. 7, F, G, C, A, B and D are neighboring prediction blocks of the current block E and are based on the order of F, G, C, A, B and D (this order may be changed, for example, the order of D, B, A, C, G and F may be used, or another order may be used, in addition, one or more neighboring prediction blocks may be deleted, for example, only F, G, C, A and D are used), the motion information associated with them (if any) is added to the motion compensation candidate list in the order List0 as a priority, and a check for duplicate entries must be performed in the adding process until the detection of all the motion information is completed or the length of the motion compensation candidate list reaches X-1.

[000328] In step 2, since the length of the motion compensation candidate list is less than X, the unidirectional motion information pointing to List0 (if any) in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed during the adding process.

[000329] In step 3, if the length of the motion compensation candidate list is less than X, the unidirectional motion information indicating List1 (if any) in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000330] In step 4, if the length of the motion compensation candidate list is less than X, a repeated filling operation is performed for the last unidirectional motion information in the motion compensation candidate list until the length of the motion compensation candidate list becomes X.

[000331] In the above embodiment of the invention, X may be any positive integer, for example, the value of X may be 4 or 5, etc.

[000332] Application scenario 2: In step 1, F, G, C, A, B, and D are neighboring prediction blocks of the current block E, and depending on the order of F, G, C, A, B, and D, their corresponding motion information (if any) is added to the motion compensation candidate list in the order of List0 as a priority, and a check for duplicate entries must be performed in the adding process until the detection of all the motion information is completed or until the length of the motion compensation candidate list reaches X.

[000333] In step 2, if the length of the motion compensation candidate list is less than X, the unidirectional motion information pointing to List0 (if any) in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000334] In step 3, if the length of the motion compensation candidate list is less than X, the unidirectional motion information indicating List1 (if any) in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000335] In step 4, if the length of the motion compensation candidate list is less than X, a repeated filling operation is performed for the last unidirectional motion information in the motion compensation candidate list until the length of the motion compensation candidate list becomes X.

[000336] In the above embodiment of the invention, X may be any positive integer, for example, the value of X may be 4 or 5, etc.

[000337] For example, application scenario 1 differs from application scenario 2 in that application scenario 1 has at least one reserved position for temporary motion information, that is, in step 1, the length of the motion compensation candidate list reaches a maximum of X-1, rather than X.

[000338] Application scenario 3: In step 1, F, G, C, A, B, and D are neighboring prediction blocks of the current block E, and based on the order of F, G, C, A, B, and D, the unidirectional motion information (if any) is added to the motion compensation candidate list in turn, and a check for duplicate entries must be performed in the adding process until all the unidirectional motion information is detected or until the length of the motion compensation candidate list reaches X-1.

[000339] In step 2, if the length of the motion compensation candidate list is less than X-1, the bidirectional motion information (if any) of F, G, C, A, B, and D is added to the motion compensation candidate list in the order of List0 as a priority, and a check for duplicate entries must be performed during the adding process until all the bidirectional motion information is detected or until the length of the motion compensation candidate list reaches X-1.

[000340] In step 3, if the length of the motion compensation candidate list is less than X, the unidirectional motion information (if any) pointing to List0 in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000341] In step 4, if the length of the motion compensation candidate list is less than X, the unidirectional motion information (if any) pointing to List1 in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000342] In step 5, if the length of the motion compensation candidate list is less than X, a repeated filling operation is performed for the last unidirectional motion information in the motion compensation candidate list until the length of the motion compensation candidate list becomes X.

[000343] In the above embodiment of the invention, X may be any positive integer, for example, the value of X may be 4 or 5, etc.

[000344] In a possible implementation, scenario 3 of use may be described in another form:

[000345] In step 1, F, G, C, A, B, and D are neighboring prediction blocks of the current block E, and depending on the order of F, G, C, A, B, and D, the availability of F, G, C, A, B, and D is determined: if F is present and the inter prediction mode is adopted, it may be determined that F is available; otherwise, F is determined as unavailable. If G is present and the inter prediction mode is adopted, it may be determined that G is available; otherwise, G is determined as unavailable. If C is present and the inter prediction mode is adopted, it may be determined that C is available; otherwise, C is determined as unavailable. If A is present and the inter prediction mode is adopted, it may be determined that A is available; otherwise, A is determined as unavailable. If B is present and the inter prediction mode is adopted, it may be determined that B is available; otherwise, B is determined as unavailable. If D is present and the inter prediction mode is adopted, it may be determined that D is available; otherwise, D is determined as unavailable.

[000346] In step 2, according to the order of F, G, C, A, B, and D, the available unidirectional motion information is added to the motion compensation candidate list in turn, and a check for duplicate entries must be performed during the adding process until the length of the motion compensation candidate list reaches X-1 or until the traversal is completed.

[000347] In step 3, if the length of the motion compensation candidate list is less than X-1, the available bidirectional motion information is divided into unidirectional motion information pointing to List0 and unidirectional motion information pointing to List1, which are then added to the motion compensation candidate list in turn and checked for duplicate entries until the length of the motion compensation candidate list becomes X-1 or the bypass is completed.

[000348] In step 4, the obtained bidirectional motion information is divided into unidirectional motion information pointing to List0 and unidirectional motion information pointing to List1; a check for duplicate entries is performed first for the unidirectional motion information, and if the unidirectional motion information is not repeated, the unidirectional motion information is added to the motion compensation candidate list until the length of the motion compensation candidate list becomes X or the bypass is completed.

[000349] In step 5, if the length of the motion compensation candidate list is less than X, a repeated filling operation is performed for the last unidirectional motion information in the motion compensation candidate list until the length of the motion compensation candidate list becomes X.

[000350] In the above embodiment of the invention, X may be any positive integer, for example, the value of X may be 4 or 5, etc.

[000351] Application scenario 4: In step 1, F, G, C, A, B, and D are neighboring prediction blocks of the current block E, and depending on the order of F, G, C, A, B, and D, the unidirectional motion information (if any) is added to the motion compensation candidate list in turn, and a check for duplicate entries must be performed during the adding process until all the unidirectional motion information is detected or until the length of the motion compensation candidate list reaches X.

[000352] In step 2, if the length of the motion compensation candidate list is less than X, the bidirectional motion information (if any) of F, G, C, A, B, and D is added to the motion compensation candidate list in the order of List0 as a priority, and a check for duplicate entries must be performed in the adding process until all the bidirectional motion information is detected or until the length of the motion compensation candidate list reaches X.

[000353] In step 3, if the length of the motion compensation candidate list is less than X, the unidirectional motion information (if any) pointing to List0 in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000354] In step 4, if the length of the motion compensation candidate list is less than X, the unidirectional motion information (if any) pointing to List1 in the obtained temporal motion information is added to the motion compensation candidate list, and a check for duplicate entries must be performed in the adding process.

[000355] In step 5, if the length of the motion compensation candidate list is less than X, a repeated filling operation is performed for the last unidirectional motion information in the motion compensation candidate list until the length of the motion compensation candidate list becomes X.

[000356] In the above embodiment of the invention, X may be any positive integer, for example, the value of X may be 4 or 5, etc.

[000357] For example, application scenario 3 differs from application scenario 4 in that application scenario 3 has at least one reserved position for temporary motion information, that is, in step 1, the length of the motion compensation candidate list reaches a maximum of X-1, but does not reach X.

[000358] In the above application scenarios 1-4, the circuit for obtaining temporal information about the motion relates to related technical fields, for example, temporal information about the motion of the current block can be obtained through a combined block of a combined image, which is not limited here.

[000359] Application scenario 5: In application scenarios 1 to 4, if the length of the motion compensation candidate list is less than X, a repeated filling operation is performed for the last unidirectional motion information in the motion compensation candidate list. In application scenario 5, the repeated filling operation is no longer performed for the last unidirectional motion information, and the filling operation is performed by calculation. For example, with any valid motion information (x, y, ref_idx, ListX) in the motion compensation candidate list as a basis, ref_idx and ListX represent a reference index and a reference list, respectively, which are collectively referred to as reference image information, and at least one possible motion information can be added as follows:

[000360] (x+a, y+b, ref_idx, ListX), where a and b may be any integer; (k1 * x, k1 * y, ref_idx new1, ListX), where k1 is any positive integer other than zero, that is, a scaling operation is performed on the motion vector; (k2 * x, k2 * y, ref_idx_new2, ListY), where k2 is any positive integer other than zero, that is, a scaling operation is performed on the motion vector. In addition, addition of zero motion vectors may also be performed, that is, the possible motion information may be (0, 0, ref_idx3, ListZ). For the above possible motion information, when the possible motion information is added to the motion compensation candidate list, a check for duplicate entries may or may not be performed.

[000361] Application scenario 6: In application scenarios 1 to 4, the length of the motion compensation candidate list is X. For embodiment 9 of the invention, two possible motion information is selected from the motion compensation candidate list as the target motion information of the current block. Therefore, the length of the motion compensation candidate list may be from 2 to H, where the value of H may be configured based on experience, for example, configured as 4 or 5, etc. Based on this, the encoder side can set the length of the motion compensation candidate list (for example, the length of the motion compensation candidate list in the AWP mode or the length of the motion compensation candidate list in the advanced angular weighted prediction (EAWP) mode) through the sequence level syntax. For example, awpmaxcandnum is used to specify the length of the motion compensation candidate list. If the value specified by awp_max_cand_num is "a", which is in the range of 2 to H, the decoder side can determine the length of the motion compensation candidate list based on awp_max_cand_num, for example, the length of X is "a". Optionally, awp_max_cand_num_minus2 is used to specify the length of the motion compensation candidate list. If the value specified by awp_max_cand_num_minus2 is "a", which is in the range of 0 to (H-2), the decoder side can determine the length of the motion compensation candidate list based on awp_max_cand_num_minus2, for example, the length of X is a+2. Optionally, the encoder side may specify the length of the motion compensation candidate list through the picture level syntax, for example, the length of the motion compensation candidate list is specified based on slice_awp_max_cand_num. If the value specified by slice_awp_max_cand_num is "a", which is in the range from 2 to H, the decoder side may determine the length of the motion compensation candidate list based on slice_awp_max_cand_num, for example, the length of X is "a". Optionally, slice_awp_max_cand_num_minus2 is used to specify the length of the motion compensation candidate list. If the value specified by slice_awp_max_cand_num_minus2 is "a", which is in the range from 0 to (H-2), the decoder side may determine the length of the motion compensation candidate list based on slice_awp_max_cand_num_minus2, for example, the length of X is a+2.

[000362] Application scenario 7: In the application scenarios 14, a candidate list for motion compensation may be created. In the application scenario 7, the candidate list for motion compensation is designated as AwpUniArray, and the motion information in the AwpUniArray is referred to as possible motion information. For embodiment 9 of the invention, the possible motion information may be selected from the AwpUniArray as the target motion information of the current block, and then a pixel position prediction value is determined based on the target motion information.

[000363] For embodiment 9 of the invention, two possible motion information are selected from the AwpUniArray as the first target motion information and the second target motion information of the current block. Based on this, the decoder side can analyze AwpCandIdx0 and AwpCandIdx1 from the coded bitstream and assign the (AwpCandIdx0+1)-th motion information in the AwpUniArray to mvAwp0L0, mvAwp0L1, RefIdxAwp0L0 and RefIdxAwp0L1, and assign the (AwpCandIdx1+1)-th motion information in the AwpUniArray to mvAwp1L0, mvAwp1L1, RefIdxAwp1L0 and RefIdxAwp1L1. The (AwpCandIdx0+1)th motion information in AwpUniArray can also be assigned to mvAwp1L0, mvAwp1L1, RefIdxAwp1L0, and RefIdxAwp1L1, and the (AwpCandIdx1+1)th motion information in AwpUniArray can be assigned to mvAwp0L0, mvAwp0LxLd0, RefIdxAwp0L0, and RefIdxAwp0L1, which is not limited here.

[000364] For example, AwpCandIdx0 represents the index value of the first target motion information. Therefore, the (AwpCandIdx0+1)th motion information in the AwpUniArray can be assigned to the first target motion information. For example, if the value of AwpCandIdx0 is 0, the first motion information in the AwpUniArray is assigned to the first target motion information; if the value of AwpCandIdx0 is 1, the second motion information in the AwpUniArray is assigned to the first target motion information, and so on.

[000365] mvAwp0L0, mvAwp0L1, RefIdxAwp0L0, and RefIdxAwp0L1 are combined as the first target motion information, that is, the first target motion information includes the unidirectional motion information pointing to List0 and the unidirectional motion information pointing to List1.

[000366] If the (AwpCandIdx0+1)th motion information in the AwpUniArray is unidirectional motion information pointing to List0, the first target motion information includes unidirectional motion information pointing to List0, and the unidirectional motion information pointing to List1 is zero.

[000367] If the (AwpCandIdx0+1)th motion information in the AwpUniArray is unidirectional motion information pointing to List1, the first target motion information includes unidirectional motion information pointing to List1, and the unidirectional motion information pointing to List0 is zero.

[000368] For example, mvAwp0L0 and RefIdxAwp0L0 represent one-way motion information pointing to List0 in the first target motion information, and mvAwp0L1 and RefIdxAwp0L1 represent one-way motion information pointing to List1 in the first target motion information.

[000369] If the value of RefIdxAwp0L0 is valid, this indicates that the unidirectional motion information pointing to List0 is valid. Thus, the prediction mode of the first target motion information is PRED_List0, that is, the prediction value of the pixel position of the unidirectional motion information pointing to List0 can be adopted.

[000370] If the value of RefIdxAwp0L1 is valid, this indicates that the unidirectional motion information pointing to List1 is valid. Thus, the prediction mode of the first target motion information is PRED List1, that is, the prediction value of the pixel position of the unidirectional motion information pointing to List1 can be adopted.

[000371] For example, AwpCandIdx1 represents an index value of the second target motion information. Therefore, the (AwpCandIdx1+1)th motion information in the AwpUniArray can be assigned to the second target motion information. For example, if the value of AwpCandIdx1 is 0, the first motion information in the AwpUniArray is assigned to the second target motion information; if the value of AwpCandIdx1 is 1, the second motion information in the AwpUniArray is assigned to the second target motion information, and so on.

[000372] mvAwp1L0, mvAwp1L1, RefIdxAwp1L0, and RefIdxAwp1L1 are combined as the second target motion information, that is, the second target motion information includes the unidirectional motion information pointing to List0 and the unidirectional motion information pointing to List1.

[000373] If the (AwpCandIdx1+1)th motion information in the AwpUniArray is unidirectional motion information pointing to List0, then the second target motion information includes unidirectional motion information pointing to List0, and the unidirectional motion information pointing to List1 is zero.

[000374] If the (AwpCandIdx1+1)th motion information in the AwpUniArray is unidirectional motion information pointing to List1, then the second target motion information includes unidirectional motion information pointing to List1, and the unidirectional motion information pointing to List0 is zero.

[000375] For example, mvAwp1L0 and RefIdxAwp1L0 represent one-way motion information pointing to List0 in the second target motion information, and mvAwp1L1 and RefIdxAwp1L1 represent one-way motion information pointing to List1 in the second target motion information.

[000376] If the value of RefIdxAwp1L0 is valid, this indicates that the unidirectional motion information pointing to List0 is valid. Thus, the prediction mode of the second target motion information is PRED_List0, that is, the prediction value of the pixel position of the unidirectional motion information pointing to List0 can be adopted.

[000377] If the value of RefIdxAwp1L1 is valid, this indicates that the unidirectional motion information pointing to List1 is valid. Thus, the prediction mode of the second target motion information is PRED_List1, that is, the prediction value of the pixel position of the unidirectional motion information pointing to List1 can be adopted.

[000378] Embodiment 11. In Embodiments 1 to 3 described above, it is necessary to determine a first pixel position prediction value based on a first prediction mode and to determine a second pixel position prediction value based on a second prediction mode. In this embodiment, the description is given for the first prediction mode as an inter prediction mode and the second prediction mode as an inter prediction mode as an example.

[000379] The first prediction mode is an inter prediction mode, the second prediction mode is an inter prediction mode, and a motion compensation candidate list is created, wherein the motion compensation candidate list includes at least two possible motion information. One possible motion information is selected from the motion compensation candidate list as the first source motion information of the current block, and the other possible motion information is selected from the motion compensation candidate list as the second source motion information of the current block, wherein the first source motion information and the second source motion information are different. Then, based on the first source motion information, a first target motion information of the current block is determined, and based on the second source motion information, a second target motion information of the current block is determined. For each pixel position of the current block, the first pixel position prediction value is determined based on the first target motion information, and the second pixel position prediction value is determined based on the second target motion information.

[000380] The process of creating a motion compensation candidate list may be referred to embodiments 9 and 10 of the invention and will not be repeated here. The process of determining a first prediction value based on the first target motion information and determining a second prediction value based on the second target motion information may be referred to embodiment 9 of the invention and will not be repeated here. Unlike embodiment 9 of the invention, in embodiment 11 of the invention, the possible motion information selected from the motion compensation candidate list is the original motion information, not the target motion information. After the original motion information is obtained, the target motion information may be obtained based on the original motion information. The specific obtaining process may be referred to subsequent embodiments of the invention.

[000381] Thus, for the above-described case where at least one of the first prediction mode or the second prediction mode is an inter prediction mode, a motion compensation candidate list may be created, wherein the motion compensation candidate list includes at least one possible motion information. Then, two possible motion information are selected from the motion compensation candidate list as the first source motion information and the second source motion information of the current block. The first target motion information is determined based on the first source motion information, and the second target motion information is determined based on the second source motion information. Then, two pixel position prediction values are determined based on these two instances of the target motion information.

[000382] To determine the target motion information based on the original motion information, embodiments of the invention provide schemes for superimposing the unidirectional motion information on the motion vector difference. For example, the original motion information includes the original motion vector, and the target motion information includes the target motion vector. To determine the target motion vector of the current block based on the original motion vector, a motion vector difference (MVD) corresponding to the original motion vector can be obtained, and the target motion vector is determined based on the motion vector difference and the original motion vector, that is, the sum of the motion vector difference and the original motion vector is taken as the target motion vector.

[000383] Embodiment 12. Based on Embodiment 11, direction information and amplitude information of the motion vector difference are matched. If the direction information indicates a rightward direction and the amplitude information indicates that the amplitude is Ar, the motion vector difference is (Ar, 0). If the direction information indicates a downward direction and the amplitude information indicates that the amplitude is Ad, the motion vector difference is (0, -Ad). If the direction information indicates a leftward direction and the amplitude information indicates that the amplitude is Al, the motion vector difference is (-Al, 0). If the direction information indicates an upward direction and the amplitude information indicates that the amplitude is Au, the motion vector difference is (0, Au). If the direction information indicates a right-upward direction and the amplitude information indicates that the amplitude is Ar, the motion vector difference is (Aru, Aru). If the direction information indicates the left-up direction and the amplitude information indicates the amplitude is Alu, the motion vector difference is (-Alu, Alu). If the direction information indicates the left-down direction and the amplitude information indicates the amplitude is Aid, the motion vector difference is (-Ald, -Ald). If the direction information indicates the right-down direction and the amplitude information indicates the amplitude is Ard, the motion vector difference is (Ard, -Ard).

[000384] Note that the above amplitudes Ar, Ad, Al, Au, Aru, Alu, Ald and Ard respectively indicate one set of values, and the values in the sets of amplitudes of different directions may be completely the same or partially the same, or completely different.

[000385] For example, the motion vector difference may support all or part of the above-mentioned directional information, the range of values of the amplitude A supported by the motion vector difference may be configured on an experimental basis, and the amplitude A may include at least one value that is not limited here.

[000386] For example, the motion vector difference supports four directions: up, down, left, and right, and the motion vector difference supports the following five step length configurations: -pixels, -pixel, 1-pixel, 2-pixel and 4-pixel, that is, the value of the amplitude A can be 1, 2, 4, 8 and 16. Thus, in the upward direction, the difference of the motion vectors can be (0, 1), (0, 2), (0, 4), (0, 8) and (0, 16). In the downward direction, the difference of the motion vectors can be (0, -1), (0, -2), (0, -4), (0, -8) and (0, -16). In the left direction, the difference of the motion vectors can be (-1.0), (-2, 0), (-4, 0), (-8, 0) and (-16, 0). In the right direction, the difference of the motion vectors can be (1, 0), (2, 0), (4, 0), (8, 0) and (16, 0).

[000387] In another example, the motion vector difference supports four directions: up, down, left, and right, and the motion vector difference supports the following six step length configurations: -pixels, -pixel, 1-pixel, 2-pixel, 3-pixel and 4-pixel, that is, the value of the amplitude A can be equal to 1, 2, 4, 8 and 16. Thus, in the upward direction, the difference in the motion vectors can be (0, 1), (0, 2), (0, 4), (0, 8), (0, 12) and (0, 16). The differences in the motion vectors of other directions can be realized in the same way as in the “up” direction, and are not repeated here.

[000388] In another example, the motion vector difference supports eight directions: up, down, left, right, left-up, left-down, right-up, and right-down, and the motion vector difference supports the following three step length configurations: -pixels, -pixel and 1-pixel, that is, the value of the amplitude A can be equal to 1, 2 and 4. Thus, in the left-up direction, the difference of the motion vectors can be (-1, 1), (-2, 2) and (-4, 4). When the direction is right-up, the difference of the motion vectors can be (1, 1), (2, 2) and (4, 4). The differences of the motion vectors of other directions can be realized in the same way as in the directions "left up and right up", and the values of the differences of the motion vectors can be specified in the above convention and will not be repeated here.

[000389] In another example, the motion vector difference supports four directions: up, down, left, and right, and the motion vector difference supports the following four step length configurations: -pixels, -pixel, 1-pixel and 2-pixel, that is, the amplitude value A can be equal to 1, 2, 4 and 8.

[000390] The above examples are only illustrative, and no limitations are made herein. For example, the direction supported by the motion vector difference may be arbitrarily selected, such as up, down, left, right, left-up and left-down, or two directions such as up and down, etc. As another example, the step length configuration supported by the motion vector difference may be changeable and can be flexibly configured. In another example, the step length configuration may be adaptively configured based on encoding parameters such as QP (quantization parameter) and the like. For example, 1-pixel, 2-pixel, 4-pixel and 8-pixel may be adopted for a larger QP, and -pixels, -pixel, 1-pixel and 2-pixel can be adopted for a smaller QP. In another example, a suitable step length configuration can be set at a sequence level, a picture level, a frame level, a slice level, a tile level, a patch level, a CTU level, etc., and the decoder side performs a decoding operation according to the step length configurations analyzed at a sequence level, a picture level, a frame level, a slice level, a tile level, a patch level, a CTU level, etc.

[000391] For ease of description, in the following embodiments of the invention, the motion vector difference supports up and down directions and supports step length configurations such as 1-pixel and 2-pixel in the case of accuracy -pixel motion vector difference can be (0, 4), (0, 8), (0, -4), (0, -8), that is (0, 1 << 2), (0, 1 << 3), (0, -1 << 2) and (0, -1 << 3).

[000392] After creating the motion compensation candidate list, the encoder side may sequentially traverse each possible motion vector in the motion compensation candidate list. When traversing the possible motion vector 1, the sum of the possible motion vector 1 and the motion vector difference (0, 4) is taken as the possible motion vector 1-1, and the 1-1 RDO cost value corresponding to the possible motion vector 1-1 is determined. The determination process described above is not limited here. The sum of the possible motion vector 1 and the motion vector difference (0, 8) is taken as the possible motion vector 1-2, and the 1-2 RDO cost value corresponding to the possible motion vector 1-2 is determined. The sum of the possible motion vector 1 and the motion vector difference (0, -4) is taken as the possible motion vector 1-3, and the 1-3 RDO cost value corresponding to the possible motion vector 1-3 is determined. The sum of the possible motion vector 1 and the difference of the motion vectors (0, -8) is taken as the possible 1-4 motion vector, and the 1-4 RDO cost value corresponding to the possible 1-4 motion vector is determined.

[000393] Similarly, for each of the traversed possible motion vectors, an RDO cost value may be obtained by the above-described processing. After traversing all of the possible motion vectors, a minimum RDO cost value is selected from all of the RDO cost values. If the 1-1 RDO cost value is the minimum, the encoder side may encode the following content in the encoded bitstream: an index value of the possible motion vector 1 in the motion compensation candidate list, direction information, and amplitude information of the motion vector difference (0, 4), where the direction information is used to indicate the direction of the motion vector difference (0, 4) up, and the amplitude information is used to indicate the amplitude of the motion vector difference (0, 4) and is 4. For example, the direction information indicating information may be 0, indicating the first direction in the direction list (up and down), and the amplitude information indicating information may be 4, indicating the first step length configuration in the step length configuration list (1-pixel, 2-pixel). The process described above is a simplified example, which is not limited here, provided that direction information and amplitude information can be specified.

[000394] For example, when the motion vector difference supports four directions: up, down, left, and right, and the motion vector difference supports five step length configurations: -pixels, -pixel, 1-pixel, 2-pixel and 4-pixel, the direction information of the motion vector difference can be encoded using a 2-bin fixed-length code (a total of four values), where the four values of the 2-bin fixed-length code respectively represent four directions: up, down, left and right. The amplitude information of the motion vector difference can be encoded using a truncated unary code, that is, five step length configurations are indicated by a truncated unary code.

[000395] Another example is where the motion vector difference supports four directions: up, down, left, and right, and the motion vector difference supports six step length configurations: -pixels, -pixel, 1-pixel, 2-pixel, 3-pixel and 4-pixel, the direction information of the motion vector difference can be encoded using a fixed-length 2-bin code (a total of four values), and the amplitude information of the motion vector difference can be encoded using a truncated unary code.

[000396] Another example is when the motion vector difference supports eight directions: up, down, left, right, left-up, left-down, right-up, and right-down, and the motion vector difference supports three step length configurations: -pixels, -pixel and 1-pixel, the direction information of the motion vector difference can be encoded using a 3-bin fixed-length code (eight values in total), and the amplitude information of the motion vector difference can be encoded using a truncated unary code.

[000397] In another example, when the motion vector difference supports four directions: up, down, left, right, and the motion vector difference supports four step length configurations: -pixels, -pixel, 1-pixel and 2-pixel, the direction information of the motion vector difference can be encoded using a truncated unary code, and the amplitude information of the motion vector difference can be encoded using a fixed-length 2-bin code (a total of four values).

[000398] The encoding methods described above are examples only and are not limited here.

[000399] To summarize, the encoder side may search for an optimal motion vector in a certain region, and then take the difference between the optimal motion vector and a candidate motion vector as a motion vector difference (MVD), and encode the amplitude information and the direction information of the motion vector difference and the index value of the candidate motion vector in the motion compensation candidate list into the bitstream. When the encoder side searches for an optimal motion vector in a certain region, it is expected that the direction and amplitude of the motion vector difference will be consistent, that is, the optimal motion vector is searched in limited ranges of motion vector difference, such as (Ar, 0), (0, -Ad), (-Al, 0), (0, Au), (Aru, Aru), (-Aru, Alu), (-Ald, -Ald), and (Ard, -Ard), etc., rather than in any range of motion vector difference.

[000400] After receiving the coded bitstream of the current block, the decoder side may analyze the index value of the possible motion vector in the motion compensation candidate list from the coded bitstream and select the possible motion vector corresponding to the index value in the motion compensation candidate list and take the possible motion vector as the original motion vector of the current block. The decoder side may also analyze the direction information and the amplitude information of the motion vector difference from the coded bitstream and determine the motion vector difference based on the direction information and the amplitude information.

[000401] Then, the decoder side may determine the target motion vector of the current block based on the difference of the motion vectors and the source motion vector, for example, the sum of the difference of the motion vectors and the source motion vector may be taken as the target motion of the current block.

[000402] In the above embodiment of the invention, when encoding the direction information of the motion vector difference, the encoder side may use a fixed-length code or a truncated unary code, etc. Therefore, on the decoder side, a fixed-length code or a truncated unary code, etc. may be used to decode the direction information of the motion vector difference to obtain the direction information of the motion vector difference, such as up, down, left, right, left-up, left-down, right-up, right-down, etc.

[000403] In the above embodiment of the invention, when encoding the motion vector difference amplitude information, the encoder side may use a fixed length code or a truncated unary code, etc. Therefore, the decoder side may use a fixed length code or a truncated unary code, etc. to decode the motion vector difference amplitude information to obtain motion vector difference amplitude information such as step length configurations: -pixels, -pixel, 1-pixel, 2-pixel, etc., and then determine the amplitude value A of the motion vector difference based on the step length configuration, for example -pixels, -pixel, 1-pixel and 2-pixel.

[000404] To summarize, the decoder side may analyze direction information and amplitude information of the motion vector difference from the encoded bitstream, and then determine the motion vector difference based on the direction information and the amplitude information.

[000405] In a possible implementation, the encoder side may also encode flag information in the encoded bitstream, wherein the flag information indicates superposition of the motion vector difference on the original motion vector or no superposition of the motion vector difference on the original motion vector. The flag information may be a submode flag of the EAWP mode. In addition, the EAWP mode may also be called an AWP with Motion Vector Refinement (AWP-MVR) mode. After receiving the encoded bitstream of the current block, the decoder side first analyzes the submode flag of the extended angle weighted prediction (EAWP) mode from the encoded bitstream of the current block. If the submode flag indicates superposition of the motion vector difference on the original motion vector, the decoder side analyzes direction information and amplitude information of the motion vector difference from the encoded bitstream of the current block and determines the motion vector difference based on the direction information and the amplitude information, and determines the target motion vector of the current block based on the original motion vector and the motion vector difference. If the submode flag indicates that there is no overlap between the motion vector difference and the original motion vector, the decoder side does not analyze the direction information or the amplitude information of the motion vector difference, but directly takes the original motion vector as the target motion of the current block.

[000406] For example, when the submode flag of the EAWP mode is the first submode value, this indicates that the motion vector difference is superimposed on the original motion vector; when the submode flag of the EAWP mode is the second submode value, this indicates that the motion vector difference is not superimposed on the original motion vector. The first submode value and the second submode value can be configured based on experience, for example, the first submode value can be 1, and the second submode value can be 0, or, for example, the first submode value can be 0, and the second submode value can be 1. The above are only two examples, and no limitations are made here.

[000407] In a possible implementation for embodiment 11 of the invention, the first original motion information includes a first original motion vector, the first target motion information includes a first target motion vector, the second original motion information includes a second original motion vector, and the second target motion information includes a second target motion vector. Based on the above, a first motion vector difference corresponding to the first original motion vector can be obtained, and the first target motion vector is determined based on the first motion vector difference and the first original motion vector, that is, the sum of the first motion vector difference and the first original motion vector is taken as the first target motion vector. A second motion vector difference corresponding to the second original motion vector can be obtained, and the second target motion vector is determined based on the second motion vector difference and the second original motion vector, that is, the sum of the second motion vector difference and the second original motion vector is taken as the second target motion vector.

[000408] The encoder side may use the RDO to determine a first motion vector difference corresponding to a first source motion vector and determine a second motion vector difference corresponding to a second source motion vector, which is not limited here.

[000409] When the encoder side transmits the coded bitstream of the current block to the decoder side, the coded bitstream may carry direction information and amplitude information of the first motion vector difference, as well as direction information and amplitude information of the second motion vector difference.

[000410] The decoder side, after receiving the coded bit stream of the current block, may analyze direction information and amplitude information of the first motion vector difference from the coded bit stream and determine the first motion vector difference based on the direction information and amplitude information of the first motion vector difference, and may analyze direction information and amplitude information of the second motion vector difference from the coded bit stream and determine the second motion vector difference based on the direction information and amplitude information of the second motion vector difference.

[000411] Then, the decoder side may determine a first target motion vector of the current block based on the first difference of the motion vectors and the first source motion vector, and determine a second target motion vector of the current block based on the second difference of the motion vectors and the second source motion vector.

[000412] In a possible implementation, the encoder side may also encode a first submode flag and a second submode flag of the EAWP prediction mode in the encoded bitstream, wherein the first submode flag indicates the superposition of the motion vector difference on the first source motion vector or the absence of the superposition of the motion vector difference on the first source motion vector, and the second submode flag indicates the superposition of the motion vector difference on the second source motion vector or the absence of the superposition of the motion vector difference on the second source motion vector.

[000413] In a possible implementation, the above-mentioned first submode flag may also be the first bit of a flag in the encoded bitstream, and the above-mentioned second submode flag may also be the second bit of a flag in the encoded bitstream.

[000414] The decoder side, after receiving the coded bitstream of the current block, may first analyze the first submode flag and the second submode flag of the EAWP mode from the coded bitstream of the current block. If the first submode flag indicates superposition of the motion vector difference on the first source motion vector, the decoder side analyzes the direction information and the amplitude information of the first motion vector difference from the coded bitstream and determines the first motion vector difference based on the direction information and the amplitude information of the first motion vector difference, and then determines the first target motion vector of the current block based on the first source motion vector and the first motion vector difference. If the first submode flag indicates no superposition of the motion vector difference on the first source motion vector, the decoder side does not analyze the direction information or the amplitude information of the first motion vector difference, but directly takes the first source motion vector as the first target motion vector of the current block. If the second submode flag indicates superposition of the motion vector difference on the second source motion vector, the decoder side analyzes the direction information and the amplitude information of the second motion vector difference from the encoded bit stream, and determines the second motion vector difference based on the direction information and the amplitude information of the second motion vector difference, and then determines the second target motion vector of the current block based on the second source motion vector and the second motion vector difference. If the second submode flag indicates that the motion vector difference is not superimposed on the second source motion vector, the decoder side does not analyze the direction information or the amplitude information of the second motion vector difference, but directly takes the second source motion vector as the second target motion vector of the current block.

[000415] Embodiment 13 of the invention: based on Embodiment 11 of the invention, the encoder side may search for an optimal motion vector within a certain area, then receive the difference between the optimal motion vector and the possible motion vector as a motion vector difference (MVD), and then encode information about the motion vector difference into a bitstream and encode an index value of the possible motion vector in the motion compensation candidate list into the bitstream.

[000416] When searching for an optimal motion vector in a certain area, the encoder may perform the search arbitrarily without restrictions. For example, for each possible motion vector in the candidate list for motion compensation, several boundary motion vectors are searched for by taking the possible motion vector as the center, an RDO cost value corresponding to each of the boundary motion vectors is determined, and the minimum RDO cost value is selected from all RDO cost values, and the boundary motion vector corresponding to the minimum RDO cost value is taken as the optimal motion vector.

[000417] If the optimal motion vector is obtained based on the possible motion vector 1, the possible motion vector 1 may be adopted as the original motion vector of the current block, and the difference between the optimal motion vector and the possible motion vector 1 is adopted as the motion vector difference.

[000418] The encoder side may encode the following content in the encoded bitstream: a value of the index of the candidate motion vector 1 in the motion compensation candidate list and motion vector difference information, where the motion vector difference information includes, but is not limited to, horizontal component difference information and/or vertical component difference information. The horizontal component difference information may include an absolute value of the horizontal component difference and/or a signed value of the horizontal component difference, and the vertical component difference information may include an absolute value of the vertical component difference and/or a signed value of the vertical component difference.

[000419] The decoder side may, after receiving the coded bitstream of the current block, analyze an index value of a candidate motion vector in the motion compensation candidate list from the coded bitstream, select a candidate motion vector corresponding to the index value from the motion compensation candidate list, and take the candidate motion vector as an initial motion vector of the current block. The decoder side may also analyze horizontal component difference information and/or vertical component difference information of the motion vector difference from the coded bitstream of the current block, and determine the motion vector difference based on the horizontal component difference information (for example, an absolute value of the horizontal component difference and/or a signed value of the horizontal component difference) of the motion vector difference and the vertical component difference information (for example, an absolute value of the vertical component difference and/or a signed value of the vertical component difference) of the motion vector difference.

[000420] Then, the decoder side may determine the target motion vector of the current block based on the difference of the motion vectors and the source motion vector, for example, the decoder side may take the sum of the difference of the motion vectors and the source motion vector as the target motion vector of the current block.

[000421] For example, the decoder side may determine the horizontal component difference based on the absolute value of the horizontal component difference and the signed value of the horizontal component difference, and determine the vertical vector difference based on the absolute value of the vertical component difference and the signed value of the vertical component difference. The combination of the horizontal component difference and the vertical component difference is the motion vector difference, that is, the motion vector difference includes the horizontal component difference and the vertical component difference.

[000422] In a possible implementation, the encoder side may also encode a submode flag of the EAWP mode in the encoded bitstream, wherein the submode flag indicates the superposition of the motion vector difference on the original motion vector or the absence of superposition of the motion vector difference on the original motion vector. The decoder side, after receiving the encoded bitstream of the current block, analyzes the submode flag of the EAWP mode from the encoded bitstream. If the submode flag indicates the superposition of the motion vector difference on the original motion vector, horizontal component difference information and vertical component difference information of the motion vector difference are analyzed from the encoded bitstream, the motion vector difference is determined based on the horizontal component difference information and the vertical component difference information, and the target motion vector of the current block is determined based on the original motion vector and the motion vector difference. If the submode flag indicates that there is no overlap of the motion vector difference with the source motion vector, the horizontal component difference information and the vertical component difference information of the motion vector difference are not analyzed, but the source motion vector is taken as the target motion vector of the current block.

[000423] In a possible implementation for embodiment 11 of the invention, the first original motion information includes a first original motion vector, the first target motion information includes a first target motion vector, the second original motion information includes a second original motion vector, and the second target motion information includes a second target motion vector. Based on this, a first motion vector difference corresponding to the first original motion vector can be obtained, and the first target motion vector is determined based on the first motion vector difference and the first original motion vector, that is, the sum of the first original motion vector, and the first original motion vector is taken as the first target motion vector. A second motion vector difference corresponding to the second original motion vector can be obtained, and the second target motion vector is determined based on the second motion vector difference and the second original motion vector, that is, the sum of the second motion vector difference, and the second original motion vector is taken as the second target motion vector.

[000424] For the encoder side, the coded bitstream of the current block may carry horizontal component difference information (e.g., an absolute value of the horizontal component difference and/or a signed value of the horizontal component difference) and vertical component difference information (e.g., an absolute value of the vertical component difference and/or a signed value of the vertical component difference) of the first motion vector difference, as well as horizontal component difference information and vertical component difference information of the second motion vector difference.

[000425] After receiving the coded bitstream of the current block, the decoder side may analyze the horizontal component difference information and the vertical component difference information of the first motion vector difference from the coded bitstream, and determine the first motion vector difference based on the horizontal component difference information and the vertical component difference information of the first motion vector difference. The decoder side may also analyze the horizontal component difference information and the vertical component difference information of the second motion vector difference from the coded bitstream, and determine the second motion vector difference based on the horizontal component difference information and the vertical component difference information of the second motion vector difference. The first target motion vector is determined based on the first motion vector difference and the first source motion vector, and the second target motion vector is determined based on the second motion vector difference and the second source motion vector.

[000426] In a possible implementation, the encoder side may also encode a first submode flag and a second submode flag of the EAWP mode in the encoded bitstream, wherein the first submode flag indicates the superposition of the motion vector difference on the first source motion vector or the absence of the superposition of the motion vector difference on the first source motion vector; the second submode flag indicates the superposition of the motion vector difference on the second source motion vector or the absence of the superposition of the motion vector difference on the second source motion vector.

[000427] After receiving the coded bitstream of the current block, the decoder side may first analyze the first submode flag and the second submode flag of the EAWP mode from the coded bitstream of the current block. If the first submode flag indicates that the motion vector difference is superimposed on the first source motion vector, the decoder side analyzes the horizontal component difference information and the vertical component difference information of the first motion vector difference from the coded bitstream of the current block and determines the first motion vector difference based on the horizontal component difference information and the vertical component difference information of the first motion vector difference, and then determines the first target motion vector of the current block based on the first source motion vector and the first motion vector difference. If the first submode flag indicates that the motion vector difference is not superimposed on the first source motion vector, the decoder side does not analyze the horizontal component difference information or the vertical component difference information of the first motion vector difference, but directly receives the first source motion vector as the first target motion vector of the current block. If the second submode flag indicates that the motion vector difference is superimposed on the second source motion vector, the decoder side analyzes the horizontal component difference information and the vertical component difference information of the second motion vector difference from the coded bit stream of the current block, and determines the second motion vector difference based on the horizontal component difference information and the vertical component difference information of the second motion vector difference, and then determines the second target motion vector of the current block based on the second source motion vector and the second motion vector difference. If the second submode flag indicates that the motion vector difference is not superimposed on the second source motion vector, the decoder side does not analyze the horizontal component difference information or the vertical component difference information of the second motion vector difference, but directly takes the second source motion vector as the second target motion vector of the current block.

[000428] Embodiment 14: Based on embodiments 11-13 of the invention for two motion vector differences, a corresponding syntax for superimposing a motion vector difference on unidirectional motion information is described below in combination with several specific application scenarios.

[000429] Usage Scenario 1: Table 4 shows an example of the relevant syntax, where SkipFlag indicates whether the current block is in skip mode, DirectFlag indicates whether the current block is in direct mode, and AwpFlag indicates whether the current block is in AWP mode.

[000430] awp_idx (weighted prediction angle mode index) is a value of the weighted prediction angle mode index in the skip mode or the direct mode, and the value of AwpIdx (see Embodiment 8 of the invention for determining AwpIdx) may be equal to the value of awp_idx. If there is no awp_idx in the bitstream, the value of AwpIdx is 0.

[000431] awp_cand_idx0 (first index of the angle weighted prediction mode motion information) is the first index value of the angle weighted prediction mode motion information in the skip mode or the direct mode. The value of AwpCandIdx0 (see Embodiment 10 of the invention for the corresponding definition) is equal to the value of awp_cand_idx0. If there is no awp_cand_idx0 in the bitstream, the value of AwpCandIdx0 is 0.

[000432] awp_cand_idx1 (the second index of the motion information for the angular weighted prediction mode) is the second value of the index of the motion information for the angular weighted prediction mode in the skip mode or the direct mode. The value of AwpCandIdx1 (see Embodiment 10 of the invention for the corresponding definition) is equal to the value of awp_cand_idx1. If there is no awp_cand_idx1 in the bitstream, the value of AwpCandIdx1 is 0.

[000433] awp_mvd_flag (extended angle weighted prediction mode flag) is a binary variable. When awp_mvd_flag is the first value (e.g. 1), it indicates that the current block is in the extended angle weighted prediction mode; when awp_mvd_flag is the second value (e.g. 0), it indicates that the current block is in the non-extended angle weighted prediction mode. For example, the value of AwpMvdFlag may be equal to the value of awp_mvd_flag. If there is no awp_mvd_flag in the bitstream, the value of AwpMvdFlag is 0.

[000434] awp_mvd_sub_flag0 (the first sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag0 is the first value, it indicates that the first motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag0 is the second value, it indicates that the first motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag0 may be equal to the value of awp_mvd_sub_flag0. If there is no awp_mvd_sub_flag0 in the bitstream, the value of AwpMvdSubFlag0 is 0.

[000435] awp_mvd_sub_flag1 (the second sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag1 is the first value, it indicates that the second motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag1 is the second value, it indicates that the second motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag1 may be equal to the value of awp_mvd_sub_flag1. If there is no awp_mvd_sub_flag1 in the bitstream, the following case may occur: if the value of AwpMvdFlag is 1, the value of AwpMvdSubFlag1 is 1, otherwise the value of AwpMvdSubFlag1 may be 0.

[000436] awp_mvd_dir0 (direction index value of motion vector difference for first motion information) is a direction index value of the motion vector difference for the first motion information of the extended angle weighted prediction mode. For example, the value of AwpMvdDir0 may be equal to the value of awp_mvd_dir0. If there is no awp_mvd_dir0 in the bitstream, the value of AwpMvdDir may be equal to 0.

[000437] awp_mvd_step0 (the step length index value of the motion vector difference of the first motion information) is the step length index value of the motion vector difference of the first motion information of the extended angle weighted prediction mode. For example, the value of AwpMvdStep0 may be equal to the value of awp_mvd_step0. If there is no awp_mvd_step0 in the bitstream, the value of AwpMvdStep0 may be equal to 0.

[000438] awp_mvd_dir1 (motion vector difference direction index value of the second motion information) is a motion vector difference direction index value of the second motion information of the extended angle weighted prediction mode. For example, the value of AwpMvdDir1 may be equal to the value of awp_mvd_dir1. If there is no awp_mvd_dir1 in the bitstream, the value of AwpMvdDir1 may be equal to 0.

[000439] awp_mvd_step1 (the step length index value of the motion vector difference of the second motion information) is the step length index value of the motion vector difference of the second motion information of the extended angle weighted prediction mode. For example, the value of AwpMvdStep1 may be equal to the value of awp_mvd_step1. If there is no awp_mvd_step1 in the bitstream, the value of AwpMvdStep1 may be equal to 0.

[000440] Usage Scenario 2: Table 5 shows an example of the relevant syntax, where SkipFlag indicates whether the current block is in skip mode, DirectFlag indicates whether the current block is in direct mode, and AwpFlag indicates whether the current block is in AWP mode.

[000441] For awp_idx, awp_cand_idx0 and awp_cand_idx1, reference may be made to application scenario 1 and no redundant descriptions are provided here.

[000442] awp_mvd_sub_flag0 (the first sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag0 is the first value, it indicates that the first motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag0 is the second value, it indicates that the first motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag0 may be equal to the value of awp_mvd_sub_flag0. If there is no awp_mvd_sub_flag0 in the bitstream, the value of AwpMvdSubFlag0 is 0.

[000443] awp_mvd_sub_flag1 (the second sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag1 is the first value, it indicates that the second motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag1 is the second value, it indicates that the second motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag1 may be equal to the value of awp_mvd_sub_flag1. If there is no awp_mvd_sub_flag1 in the bitstream, the value of AwpMvdSubFlag1 is 0.

[000444] For awp_mvd_dir0, awp_mvd_step0, awp_mvd_dir1 and awp_mvd_step1, you can make a reference to scenario 1 of use.

[000445] For example, application scenario 1 and application scenario 2 differ in the following: application scenario 1 has the syntax awp_mvd_flag, but application scenario 2 does not have the syntax awp_mvd_flag. In application scenario 1, the advanced angle weighted prediction mode is controlled by awp_mvd_flag, that is, the advanced angle weighted prediction mode is controlled by the master switch.

[000446] For example, application scenarios 1 and 2 may be implementations of embodiment 12 of the invention.

[000447] Usage Scenario 3: Table 6 shows an example of the relevant syntax, where SkipFlag indicates whether the current block is in skip mode, DirectFlag indicates whether the current block is in direct mode, and AwpFlag indicates whether the current block is in AWP mode.

[000448] awp_idx (weighted prediction mode corner index) is the value of the weighted prediction mode corner index in the skip mode or direct mode, and the value of AwpIdx may be equal to the value of awp_idx. If there is no awp_idx in the bitstream, the value of AwpIdx is 0.

[000449] awp_cand_idx0 (first index of motion information for angular weighted prediction mode) is the first index value of motion information for angular weighted prediction mode in the skip mode or in the direct mode. The value of AwpCandIdx0 (see embodiment 10 of the invention for the corresponding definition) is equal to the value of awp_cand_idx0. If there is no awp_cand_idx0 in the bitstream, the value of AwpCandIdx0 is 0.

[000450] awp_cand_idx1 (the second index of the motion information for the angular weighted prediction mode) is the second value of the index of the motion information for the angular weighted prediction mode in the skip mode or the direct mode. The value of AwpCandIdx1 (see embodiment 10 of the invention for the corresponding definition) is equal to the value of awp_cand_idx1. If there is no awp_cand_idx1 in the bitstream, the value of AwpCandIdx1 is 0.

[000451] awp_mvd_flag (extended angle weighted prediction mode flag) is a binary variable. When awp_mvd_flag is the first value (e.g. 1), it indicates that the current block is in the extended angle weighted prediction mode; when awp_mvd_flag is the second value (e.g. 0), it indicates that the current block is in the non-extended angle weighted prediction mode. For example, the value of AwpMvdFlag may be equal to the value of awp_mvd_flag. If there is no awp_mvd_flag in the bitstream, the value of AwpMvdFlag is 0.

[000452] awp_mvd_sub_flag0 (the first sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag0 is the first value, it indicates that the first motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag0 is the second value, it indicates that the first motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag0 may be equal to the value of awp_mvd_sub_flag0. If there is no awp_mvd_sub_flag0 in the bitstream, the value of AwpMvdSubFlag0 is 0.

[000453] awp_mvd_sub_flag1 (the second sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag1 is the first value, it indicates that the second motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag1 is the second value, it indicates that the second motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag1 may be equal to the value of awp_mvd_sub_flag1. If awp_mvd_sub_flag1 is not present in the bitstream, the following cases may occur: if the value of AwpMvdFlag is 1, the value of AwpMvdSubFlag1 is 1, otherwise the value of AwpMvdSubFlag1 may be 0. For example, awp_mv_diff_x_abs0 (the absolute value of the difference of the horizontal components of the first motion vector in the extended angular weighted prediction mode) and awp_mv_diff_y_abs0 (the absolute value of the difference of the vertical components of the first motion vector in the extended angular weighted prediction mode) are the absolute values of the first difference of motion vectors of the extended angular weighted prediction mode. The value of AwpMvDiffXAbs0 is equal to the value of awp_mv_diff_x_abs0, and the value of AwpMvDiffYAbs0 is equal to the value of awp_mv_diff_у_abs0. awp_mv_diff_x_sign0 (signed value of difference of horizontal components of first motion vector in extended angular weighted prediction mode) and awp_mv_diff_у_sign0 (signed value of difference of vertical components of first motion vector in extended angular weighted prediction mode) are the sign bits of the first motion vector difference of the extended angular weighted prediction mode. The value of AwpMvDiffXSign0 is equal to the value of awp_mv_diff_x_sign0, and the value of AwpMvDiffYSign0 is equal to the value of awp_mv_diff_у_sign0.

[000454] In a possible implementation, if the bitstream does not contain awp_mv_diff_x_sign0 or awp_mv_diff_y_sign0, the value of AwpMvDiffXSign0 or the value of AwpMvDiffYSign0 is 0. If the value of AwpMvDiffXSign0 is 0, the value of AwpMvDiffX0 is AwpMvDiffXAbs0; if the value of AwpMvDiffXSign0 is 1, then the value of AwpMvDiffX0 is -AwpMvDiffXAbs0; if the value of AwpMvDiffYSign0 is 0, then the value of AwpMvDiffY0 is AwpMvDiffYAbs0; if the value of AwpMvDiffYSign0 is 1, then the value of AwpMvDiffY0 is AwpMvDiffYAbs0. The values of AwpMvDiffX0 and AwpMvDiffY0 are in the range from -32768 to 32767.

[000455] For example, awp_mv_diff_x_abs1 (the absolute value of the difference of the horizontal components of the second motion vector in the extended angular weighted prediction mode) and awp_mv_diff_y_abs1 (the absolute value of the difference of the vertical components of the second motion vector in the extended angular weighted prediction mode) are the absolute values of the second difference of the motion vectors of the extended angular weighted prediction mode. The value of AwpMvDiffXAbs1 is equal to the value of awp_mv_diff_x_abs1, and the value of AwpMvDiffYAbs1 is equal to the value of awp_mv_diff_у_abs1. awp_mv_diff_х_sign1 (signed value of difference of horizontal components of second motion vector in extended angular weighted prediction mode) and awp_mv_diff_y_sign1 (signed value of difference of vertical components of second motion vector in extended angular weighted prediction mode) are the sign bits of the second motion vector difference of extended angular weighted prediction mode. The value of AwpMvDiffXSign1 is equal to the value of awp_mv_diff_х_sign1, and the value of AwpMvDiffYSign1 is equal to the value of awp_mv_diff_y_sign1.

[000456] In a possible implementation, if the bitstream does not contain awp_mv_diff_x_sign1 or awp_mv_diff_y_sign1, the value of AwpMvDiffXSign1 or AwpMvDiffYSign1 is 0. If the value of AwpMvDiffXSign1 is 0, AwpMvDiffX1 is equal to AwpMvDiffXAbs1; if the value of AwpMvDiffXSign1 is 1, then the value of AwpMvDiffX1 is -AwpMvDiffXAbs1. If the value of AwpMvDiffYSign1 is 0, the value of AwpMvDiffY1 may be equal to AwpMvDiffYAbsl; if the value of AwpMvDiffYSign1 is 1, the value of AwpMvDiffY1 may be equal to AwpMvDiffYAbs1. The values of AwpMvDiffX1 and AwpMvDiffY1 range from -32768 to 32767.

[000457] Usage Scenario 4: Table 7 shows an example of the relevant syntax, where SkipFlag represents whether the current block is in skip mode, DirectFlag represents whether the current block is in direct mode, and AwpFlag represents whether the current block is in AWP mode.

[000458] For awp_idx, awp_cand_idx0 and awp_cand_idx1, reference may be made to application scenario 1 and no redundant descriptions are provided here.

[000459] awp_mvd_sub_flag0 (the first sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag0 is the first value, it indicates that the first motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag0 is the second value, it indicates that the first motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag0 may be equal to the value of awp_mvd_sub_flag0. If there is no awp_mvd_sub_flag0 in the bitstream, the value of AwpMvdSubFlag0 is 0.

[000460] awp_mvd_sub_flag1 (second sub-mode flag of the advanced angle weighted prediction mode) is a binary variable. When awp_mvd_sub_flag1 is the first value, it indicates that the second motion information of the advanced angle weighted prediction mode should be superimposed on the motion information difference; when awp_mvd_sub_flag1 is the second value, it indicates that the second motion information of the advanced angle weighted prediction mode should not be superimposed on the motion information difference. For example, the value of AwpMvdSubFlag1 may be equal to the value of awp_mvd_sub_flag1. If there is no awp_mvd_sub_flag1 in the bitstream, the value of AwpMvdSubFlag1 may be equal to 0.

[000461] For example, for parameters such as awp_mv_diff_x_abs0, awp_mv_diff_y_abs0, awp_mv_diff_x_sign0, awp_mv_diff_y_sign0, awp_mv_diff_x_abs1, awp_mv_diff_y_abs1, awp_mv_diff_x_sign1, and awp_mv_diff_y_sign1 in Table 7, reference can be made to application scenario 3, and redundant descriptions are not provided here.

[000462] For example, application scenario 3 and application scenario 4 differ in the following: in application scenario 3, the syntax awp_mvd_flag is present, but in application scenario 4, the syntax awp_mvd_flag is absent. In application scenario 3, the advanced angle weighted prediction mode is controlled by awp_mvd_flag, that is, the advanced angle weighted prediction mode is controlled by the master switch.

[000463] For example, application scenario 3 and application scenario 4 are implementations of embodiment 13 of the invention.

[000464] The difference between application scenario 3 or 4 and application scenario 1 or 2 is that in application scenario 1 or 2, several preferred MVDs are used, and in application scenario 3 or 4, all MVDs within a certain search range are allowed, for example, based on the method of application scenario 5, all MVDs are searched within a certain search range.

[000465] Application Scenario 5: Table 8 shows an example of the corresponding syntax. Table 8 illustrates the encoding method of the MVD.

[000466] 1. Determine whether |MVD_X| is greater than 0, where |…| denotes the absolute value and MVD_X refers to the horizontal component of MVD.

[000467] 2. Determine whether |MVD_Y| is greater than 0, where |…| denotes the absolute value and MVD_Y refers to the vertical component of MVD.

[000468] 3. If |MVD_X| is greater than 0, determine whether |MVD_X| is greater than 1.

[000469] 4. If |MVD_Y| is greater than 0, determine whether |MVD_Y| is greater than 1.

[000470] 5. If |MVD_X| is greater than 1, encode |MVD_X| - 2.

[000471] 6. If |MVD_X| is greater than 0, encode the sign bit of MVD_X.

[000472] 7. If |MVD_Y| is greater than 1, encode |MVD_Y| - 2.

[000473] 8. If |MVD_Y| is greater than 0, encode the sign bit of MVD_Y.

[000474] Application Scenario 6: A derivative version of the above Application Scenarios 1-4 to completely merge the AWP mode and the EAWP mode, that is, the 0 range is added to the range without encoding the flag bit that should be enabled or not. For example, the motion vector difference supports four directions: up, down, left, and right, and the motion vector difference supports the following 6 step length configurations: 0-pixels, -pixels, -pixel, 1-pixel, 2-pixel, 4-pixel, that is, the configuration of the length of one step of 0-pixels is added. Based on this, Table 4 or 5 can be updated to Table 9, and Table 6 or 7 can be updated to Table 10. For the values of the corresponding syntax in Tables 9 and 10, reference can be made to Table 4 and Table 6, and redundant descriptions are not provided here.

[000475] Application Scenario 7: The above-described embodiments of the invention include an AWP mode, a DAWP (double exterior angle weighted prediction) mode, and the enabling or disabling of the EAWP mode can be controlled using a high-level syntax. For example, enabling or disabling the AWP mode is controlled using high-level syntax 1. In another example, enabling or disabling the DAWP mode is controlled using high-level syntax 2. In another example, enabling or disabling the EAWP mode is performed using high-level syntax 3. As another example, simultaneously enabling or disabling the AWP and DAWP modes is controlled using high-level syntax 4. In another example, simultaneously enabling or disabling the AWP and EAWP modes is controlled using high-level syntax 5. In another example, simultaneously enabling or disabling the DAWP and EAWP modes is controlled using high-level syntax 6. In another example, simultaneously enabling or disabling the AWP, DAWP, and EAWP modes is controlled using high-level syntax 7. The above are only examples, and no limitations are made herein.

[000476] In the above embodiments of the invention, the high-level syntax may be a high-level syntax of a sequence parameter set, or a high-level syntax of a picture parameter set, or a high-level syntax of a slice header level, or a high-level syntax of a picture header, which is not limited here. For example, based on the high-level syntax of the sequence parameter set level, by using sps_awp_enable_flag, sh_awp_enable_flag and awp_enable_flag and the like, enabling or disabling of the AWP mode can be controlled, or enabling or disabling of the EAWP mode can be controlled, or enabling or disabling of multiple modes can be controlled simultaneously, which is not limited here.

[000477] For example, in the above embodiments of the invention, the target motion information may be obtained based on the source motion information. After the target motion information is obtained, the target motion information of the current block may be stored in a manner that is not limited here.

[000478] For example, the above embodiments 1-14 of the invention may be implemented individually or in combination. For example, embodiments 1 and 2 of the invention may be combined to implement; embodiments 1 and 3 of the invention may be combined to implement; embodiments 1, 2 and 3 of the invention may be combined to implement. Embodiment 4 may be combined with one or more of embodiments 1-3 of the invention to implement; embodiment 5 of the invention (or embodiment 6 of the invention, or embodiment 7 of the invention) may be combined with embodiment 2 of the invention (or embodiment 3 of the invention) to implement. Embodiment 8 of the invention may be combined with embodiment 2 of the invention (or embodiment 3 of the invention). Embodiment 9 of the invention may be combined with one or more of embodiments 9-14 of the invention or combined with one or more of embodiments 1-8 of the invention to implement, etc. The above are merely examples, and the combinations of embodiments of the invention are not limited here.

[000479] Based on the same application concept as the above-described methods, embodiments of the present invention also provide encoding devices and decoding devices. Both types of devices can be referred to as the device shown in Fig. 8A. The device comprises an obtaining module 811, a first determining module 812, a second determining module 813, a third determining module 814, and an encoding and decoding module 815. For the encoding device, the encoding and decoding module 815 performs an encoding function, and for the decoding device, the encoding and decoding module 815 performs a decoding function.

[000480] In some examples, the encoding device comprises a module (e.g., an obtaining module 811) configured to obtain, for a current block, when it is determined that weighted prediction should be enabled for the current block, a weighted prediction angle of the current block; a module (e.g., a first determining module 812) configured to determine, for each pixel position of the current block, a surrounding corresponding position indicated by the pixel position, from surrounding positions outside the current block based on the weighted prediction angle of the current block; a module (e.g., a second determining module 813) configured to determine a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position and to determine an associated weight value of the pixel position based on the target weight value of the pixel position; a module (for example, a third determining module 814) configured to determine, for each pixel position of the current block, a first prediction value of the pixel position based on a first prediction mode of the current block and to determine a second prediction value of the pixel position based on a second prediction mode of the current block; based on the first prediction value, the target weight value, the second prediction value and the associated weight value, to determine a weighted prediction value of the pixel position; and a module (for example, the encoding and decoding module 815) configured to determine the weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

[000481] In some examples, the decoding device comprises a module (e.g., an obtaining module 811) configured to obtain, for a current block, when it is determined that weighted prediction should be enabled for the current block, a weighted prediction angle of the current block; a module (e.g., a first determining module 812) configured to determine, for each pixel position of the current block, a surrounding corresponding position indicated by the pixel position, from surrounding positions outside the current block based on the weighted prediction angle of the current block; a module (e.g., a second determining module 813) configured to determine a target weight value of the pixel position based on a reference weight value associated with the surrounding corresponding position and to determine an associated weight value of the pixel position based on the target weight value of the pixel position; a module (for example, a third determining module 814) configured to determine, for each pixel position of the current block, a first prediction value of the pixel position based on a first prediction mode of the current block and to determine a second prediction value of the pixel position based on a second prediction mode of the current block; based on the first prediction value, the target weight value, the second prediction value and the associated weight value, to determine a weighted prediction value of the pixel position; and a module (for example, the encoding and decoding module 815) configured to determine the weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

[000482] The second determining module 813 is also configured to configure reference weight values for the surrounding positions outside the current block; wherein the reference weight values of the surrounding positions outside the current block are pre-configured or configured based on the weight configuration parameters; the weight configuration parameters include a weight transformation coefficient and a weight transformation start position. For example, the weight transformation start position is determined by at least one of the following parameters: a weighted prediction angle of the current block, a weighted prediction position of the current block, or a size of the current block.

[000483] For example, the number of surrounding positions beyond the current block is determined based on the size of the current block and/or the weighted prediction angle of the current block; the reference weights of the surrounding positions beyond the current block monotonically increase or monotonically decrease.

[000484] For example, the reference weight values of the surrounding positions outside the current block include one or more reference weight values of the target positions, one or more reference weight values of the first neighboring positions of the target positions, and one or more reference weight values of the second neighboring positions of the target positions. Optionally, all of the reference weight values of the first neighboring positions are second reference weight values, and all of the reference weight values of the second neighboring positions are third reference weight values, wherein the second reference weight values are different from the third reference weight values.

[000485] The target positions include one reference weight value or at least two reference weight values; if the target positions include at least two reference weight values, said at least two reference weight values of the target positions monotonically increase.

[000486] For example, the weighted prediction angle of the current block is a horizontal angle; or the weighted prediction angle of the current block is a vertical angle; or the absolute value of the slope of the weighted prediction angle of the current block is equal to the n-th power of 2, where n is an integer.

[000487] For example, surrounding positions beyond the current block may include one or more whole pixel positions or one or more sub-pixel positions, or both one or more whole pixel positions and one or more sub-pixel positions. Surrounding positions beyond the current block may include: surrounding positions in one row above the current block or surrounding positions in one column to the left of the current block.

[000488] The second determining module 813 is configured to determine, if the surrounding corresponding position is an integer pixel position, and the integer pixel position is set with a reference weight value, a target weight value of the pixel position based on the reference weight value of the integer pixel position; alternatively, if the surrounding corresponding position is a sub-pixel position, and the sub-pixel position is set with a reference weight value, determine the target weight value of the pixel position based on the reference weight value of the sub-pixel position.

[000489] For example, the first prediction mode is an inter prediction mode, and the second prediction mode is an inter prediction mode.

[000490] For example, the third determining module 814 is also configured to create a motion compensation candidate list, wherein the motion compensation candidate list includes at least two information about possible motion; select one information about possible motion from the motion compensation candidate list as the first target motion information of the current block; determine a first pixel position prediction value based on the first target motion information; select another information about possible motion from the motion compensation candidate list as the second target motion information of the current block; determine a second pixel position prediction value based on the second target motion information.

[000491] The third determining module 814 is also configured to create a motion compensation candidate list, wherein the motion compensation candidate list includes at least two pieces of possible motion information; select one possible motion information from the motion compensation candidate list as the first source motion information of the current block; determine a first target motion information of the current block based on the first source motion information; determine a first pixel position prediction value based on the first target motion information; select another possible motion information from the motion compensation candidate list as the second source motion information of the current block; determine a second target motion information of the current block based on the second source motion information; determine a second pixel position prediction value based on the second target motion information.

[000492] The first source motion information includes a first source motion vector, the first target motion information includes a first target motion vector, the second source motion information includes a second source motion vector, and the second target motion information includes a second target motion vector. The third determining module 814 is also configured to: obtain a motion vector difference corresponding to the first source motion vector, and determine the first target motion vector based on the motion vector difference corresponding to the first source motion vector and the first source motion vector; obtain a motion vector difference corresponding to the second source motion vector, and determine the second target motion vector based on the motion vector difference corresponding to the second source motion vector and the second source motion vector. If the device is applied to the decoder side, the third determining module 814 is also configured to analyze direction information and amplitude information of the corresponding motion vector difference from the coded bitstream of the current block, and determine the corresponding motion vector difference based on the direction information and the amplitude information of the corresponding motion vector difference.

[000493] For example, if the direction information indicates a right direction and the amplitude information indicates that the amplitude is Ar, the motion vector difference is (Ar, 0); if the direction information indicates a downward direction and the amplitude information indicates that the amplitude is Ad, the motion vector difference is (0, -Ad); if the direction information indicates a left direction and the amplitude information indicates that the amplitude is Al, the motion vector difference is (-Al, 0); if the direction information indicates an upward direction and the amplitude information indicates that the amplitude is Au, the motion vector difference is (0, Au).

[000494] The third determining module 814 is also configured to analyze flag information from the coded bit stream of the current block; if the flag information indicates a superposition of the motion vector difference on the original motion vector, analyze the direction information and the amplitude information of the corresponding motion vector difference from the coded bit stream of the current block.

[000495] Based on the idea of applying the above-mentioned methods, embodiments of the present invention provide a device on the decoder side (also called a video decoder). In terms of hardware, the hardware architecture is shown in Fig. 8B. The device on the decoder side comprises one or more processors 821 and a computer-readable storage medium 822, wherein the computer-readable storage medium 822 stores computer-readable instructions executable by the processor 821. The processor 821 is configured to execute the computer-readable instructions for performing the methods described above. For example, the processor 821 is configured to execute the computer-readable instructions for performing the following steps: when it is determined that weighted prediction should be enabled for the current block, obtaining a weighted prediction angle of the current block for the current block; for each pixel position of the current block, determining a surrounding corresponding position indicated by the pixel position from the surrounding positions outside the current block based on the weighted prediction angle of the current block; determining a target weight value of a pixel position based on a reference weight value associated with a surrounding corresponding position, and determining an associated weight value of the pixel position based on the target weight value of the pixel position; for each pixel position of a current block, determining a first prediction value of the pixel position based on a first prediction mode of the current block and determining a second prediction value of the pixel position based on a second prediction mode of the current block; based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determining a weighted prediction value of the pixel position; determining the weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

[000496] Based on the same application concept as the above-described methods, embodiments of the present invention provide an encoder-side device (also referred to as a video encoder). In terms of hardware, the hardware architecture is shown in Fig. 8C. The encoder-side device comprises one or more processors 831 and a computer-readable storage medium 832 that stores computer-readable instructions executable by the processor 831. The processor 831 is configured to execute the computer-readable instructions for performing the methods described above. For example, the processor 831 is configured to execute the computer-readable instructions for performing the following: when it is determined that weighted prediction should be enabled for the current block, obtaining a weighted prediction angle of the current block for the current block; for each pixel position of the current block, determining a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle of the current block; determining a target weight value of a pixel position based on a reference weight value associated with a surrounding corresponding position, and determining an associated weight value of the pixel position based on the target weight value of the pixel position; for each pixel position of a current block, determining a first prediction value of the pixel position based on a first prediction mode of the current block and determining a second prediction value of the pixel position based on a second prediction mode of the current block; based on the first prediction value, the target weight value, the second prediction value and the associated weight value, determining a weighted prediction value of the pixel position; determining the weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block.

[000497] Based on the same application concept as in the above-described methods, embodiments of the present invention also provide a camera device that may include the encoding and decoding device in any of the above-mentioned embodiments of the invention. The camera device may use the processes described above.

[000498] Based on the same application concept as in the above-described methods, embodiments of the present invention provide a computer-readable storage medium storing several computer instructions that are executed by a processor to perform the methods disclosed in the above-described embodiments of the present invention, for example, the encoding and decoding methods in the above-described embodiments of the invention.

[000499] The systems, devices, modules or units described in the above embodiments of the invention may be implemented as computer chips or objects or as products with certain functions. A typical implementation device is a computer, and the computer, in a specific form, may be a personal computer, a portable computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation equipment, an e-mail transceiver, a game console, a tablet computer, a wearable equipment or a combination of any of these devices. For convenience of description, the above-mentioned devices are divided into various blocks according to functions in the description. When implementing the present invention, the functions of the various blocks can be implemented in software and/or hardware.

[000500] It will be understood by those skilled in the art that embodiments of the present invention may be provided in the form of methods, systems, or computer program products. Accordingly, the present invention may take the form of a purely hardware embodiment, a purely software embodiment, or an embodiment combining both software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-accessible storage media (including, but not limited to, disk storage devices, CD-ROMs, optical storage devices, etc.) that contain computer-accessible program codes.

[000501] The above description presents only some examples of the implementation of the present invention and does not limit the invention. Those skilled in the art will understand that the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. of the present invention are within the scope of the claims.

Claims (130)

1. A decoding method when it is determined that weighted prediction should be enabled for the current block, comprising: get the weighted prediction angle of the current block; for each pixel position of the current block determining a surrounding corresponding position specified by the pixel position from surrounding positions outside the current block based on the weighted prediction angle; determining a target weight value of a pixel position based on a reference weight value associated with a surrounding corresponding position; determining an associated pixel position weight value based on a target pixel position weight value; for each pixel position of the current block determining a first pixel position prediction value based on a first prediction mode of the current block; determining a second pixel position prediction value based on a second prediction mode of the current block; and determining a weighted prediction value of a pixel position based on the first prediction value, the target weight value, the second prediction value, and the associated weight value; and determining the weighted prediction values of the current block based on the weighted prediction values of all pixel positions of the current block, wherein the first prediction mode is an inter prediction mode, and the second prediction mode is an inter prediction mode, and the method further comprises: creating a motion compensation candidate list, wherein the motion compensation candidate list comprises at least two pieces of possible motion information, wherein determining the first pixel position prediction value based on the first prediction mode of the current block and determining the second pixel position prediction value based on the second prediction mode of the current block includes: analyzing the first flag information from the encoded bitstream of the current block; and, if it is determined that the first flag information indicates superposition of the motion vector difference on the first original motion vector, selecting one possible motion information from the motion compensation candidate list as the first original motion information of the current block; determining a first target information about the movement of the current block based on the first source information about the movement of the current block; and determining a first pixel position prediction value based on the first target motion information; and analyzing the second flag information from the encoded bitstream of the current block; and, if it is determined that the second flag information indicates the superposition of the motion vector difference on the second original motion vector, selecting other possible motion information from the list of candidates for motion compensation as the second original motion information of the current block; determining a second target information about the motion of the current block based on the second source information about the motion of the current block; and determining a second pixel position prediction value based on the second target motion information. 2. The method according to item 1, in which the first source motion information comprises a first source motion vector, the first target motion information comprises a first target motion vector, and determining the first target motion information of the current block based on the first source motion information includes: obtaining the difference of the motion vectors corresponding to the first initial motion vector; and determining a first target motion vector based on the difference between the motion vectors corresponding to the first source motion vector and the first source motion vector; wherein the second source motion information comprises a second source motion vector, the second target motion information comprises a second target motion vector, and determining the second target motion information of the current block based on the second source motion information includes: obtaining the difference in motion vectors corresponding to the second original motion vector; and determining a second target motion vector based on the difference between the motion vectors corresponding to the second source motion vector and the second source motion vector. 3. The method according to item 2, wherein obtaining the difference in motion vectors corresponding to the first initial motion vector includes: analyzing the direction information and the amplitude information of the motion vector difference corresponding to the first source motion vector from the encoded bitstream of the current block; and determining the difference in motion vectors corresponding to the first initial motion vector based on the direction information and the amplitude information of the difference in motion vectors corresponding to the first initial motion vector, and obtaining the difference in motion vectors corresponding to the second original motion vector includes: analyzing the direction information and the amplitude information of the motion vector difference corresponding to the second original motion vector from the encoded bitstream of the current block; and determining the difference in motion vectors corresponding to the second original motion vector based on the direction information and the amplitude information of the difference in motion vectors corresponding to the second original motion vector. 4. The method according to paragraph 3, in which, if it is determined that the direction information indicates the right direction and the amplitude information indicates that the amplitude is Ar, the motion vector difference is (Ar, 0); if it is determined that the direction information indicates the downward direction and the amplitude information indicates that the amplitude is Ad, the motion vector difference is (0, -Ad); if it is determined that the direction information indicates the left direction and the amplitude information indicates that the amplitude is Al, the motion vector difference is (-Al, 0); and if it is determined that the direction information indicates the upward direction and the amplitude information indicates that the amplitude is Au, the motion vector difference is (0, Au). 5. The method according to claim 1, wherein determining the first prediction value of the pixel position based on the first prediction mode of the current block and determining the second prediction value of the pixel position based on the second prediction mode of the current block includes: analyzing the first flag information from the encoded bitstream of the current block; and, if it is determined that the first flag information indicates that there is no overlap of the motion vector difference with the first source motion vector, selecting one possible motion information from the motion compensation candidate list as the first target motion information of the current block; and determining a first pixel position prediction value based on the first target motion information; and analyzing the second flag information from the encoded bitstream of the current block; and, if it is determined that the second flag information indicates that there is no overlap of the motion vector difference with the second source motion vector, selecting other possible motion information from the motion compensation candidate list as the second target motion information of the current block; and determining a second pixel position prediction value based on the second target motion information. 6. The method according to claim 1, wherein before determining the target weight value of the pixel position based on the reference weight value associated with the surrounding corresponding position, the method also includes: configuring reference weight values for surrounding positions outside the current block, wherein the reference weight values of the surrounding positions outside the current block are configured based on the weight configuration parameters, and the weight configuration parameters include a weight conversion factor and a weight conversion start position, wherein the initial position of the weight transformation is determined by at least one of the following parameters: a weighted prediction angle of the current block, a weighted prediction position of the current block, or a size of the current block. 7. The method according to item 6, in which surrounding positions outside the current block contain surrounding positions one row above the current block or surrounding positions one column to the left of the current block, wherein the number of surrounding positions beyond the current block is determined based on the size of the current block and/or the weighted prediction angle of the current block, and the reference weight values of surrounding positions outside the current block monotonically increase or monotonically decrease. 8. The method according to item 6, in which the reference weight values of the surrounding positions outside the current block include one or more reference weight values of the target positions, one or more reference weight values of the first adjacent positions of the target positions and one or more reference weight values of the second adjacent positions of the target positions, wherein all of said one or more reference weight values of the first adjacent positions are second reference weight values, all of said one or more reference weight values of the second adjacent positions are third reference weight values, and the second reference weight values differ from the third reference weight values; the target positions include one reference weight value or at least two reference weight values; and, if it is determined that the target positions include at least two reference weight values, these at least two reference weight values of the target positions monotonically increase or monotonically decrease. 9. The method according to any one of paragraphs 1-8, in which the weighted prediction angle of the current block is the horizontal angle; or the weighted prediction angle of the current block is the vertical angle; or The absolute value of the slope of the weighted prediction angle of the current block is equal to the n-th power of 2, where n is an integer. 10. The method according to any one of claims 1-8, wherein the surrounding positions outside the current block include one or more integer pixel positions, or one or more sub-pixel positions, or both one or more integer pixel positions and one or more sub-pixel positions; wherein determining the target weight value of the pixel position based on the reference weight value associated with the surrounding corresponding position includes: if it is determined that the surrounding corresponding position is an integer pixel position, and the integer pixel position is set with a reference weight value, determining a target weight value of the pixel position based on the reference weight value of the integer pixel position; and, if it is determined that the surrounding corresponding position is a sub-pixel position, and the sub-pixel position is set with a reference weight value, determining a target weight value of the pixel position based on the reference weight value of the sub-pixel position. 11. A decoding device used on the decoding side and comprising: an obtaining module configured to obtain a weighted prediction angle of a current block when it is determined that weighted prediction should be enabled for the current block; a first determining module configured to determine, for each pixel position of the current block, a surrounding corresponding position indicated by the pixel position from surrounding positions outside the current block based on the weighted prediction angle; a second determining module configured to determine a target weight value of a pixel position based on a reference weight value associated with a surrounding corresponding position, and to determine an associated weight value of a pixel position based on the target weight value of the pixel position; a third determining module configured to, for each pixel position of the current block, determine a first prediction value of the pixel position based on a first prediction mode of the current block, determine a second prediction value of the pixel position based on a second prediction mode of the current block; and determine a weighted prediction value of the pixel position based on the first prediction value, the target weight value, the second prediction value and the associated weight value; and an encoding and decoding module configured to determine weighted prediction values of a current block based on weighted prediction values of all pixel positions of the current block; wherein the first prediction mode is an inter prediction mode, the second prediction mode is an inter prediction mode, and the third determining module is also configured to: creating a list of candidates for motion compensation, wherein the list of candidates for motion compensation contains at least two pieces of information about possible motion, wherein when determining the first value of the pixel position prediction based on the first prediction mode of the current block and determining the second value of the pixel position prediction based on the second prediction mode of the current block, the third determining module is configured to: analyzing first flag information from an encoded bitstream of the current block and, if it is determined that the first flag information indicates a superposition of the motion vector difference on the first source motion vector, selecting one possible motion information from a motion compensation candidate list as the first source motion information of the current block; determining a first target motion information of the current block based on the first source motion information of the current block; and determining a first pixel position prediction value based on the first target motion information; and analyzing the second flag information from the coded bitstream of the current block and, if it is determined that the second flag information indicates superposition of the motion vector difference on the second source motion vector, selecting other possible motion information from the motion compensation candidate list as the second source motion information of the current block; determining a second target motion information of the current block based on the second source motion information of the current block; and determining a second pixel position prediction value based on the second target motion information. 12. The device according to item 11, in which the first source motion information comprises a first source motion vector, the first target motion information comprises a first target motion vector, and, when determining the first target motion information of the current block based on the first source motion information, the third determining module is configured to: obtaining the difference of the motion vectors corresponding to the first initial motion vector; and determining a first target motion vector based on the difference between the motion vectors corresponding to the first source motion vector and the first source motion vector; wherein the second source motion information comprises a second source motion vector, the second target motion information comprises a second target motion vector, and, when determining the second target motion information of the current block based on the second source motion information, the third determining module is configured to: obtaining the difference in motion vectors corresponding to the second initial motion vector; and determining a second target motion vector based on the difference between the motion vectors corresponding to the second source motion vector and the second source motion vector. 13. The device according to item 12, in which, upon receiving the difference in motion vectors corresponding to the first initial motion vector, the third determination module is configured to: analyzing the direction information and the amplitude information of the motion vector difference corresponding to the first original motion vector from the encoded bitstream of the current block; and determining the difference in motion vectors corresponding to the first initial motion vector based on the information about the direction and the information about the amplitude of the difference in motion vectors corresponding to the first initial motion vector, and, upon receiving the difference in motion vectors corresponding to the second initial motion vector, the third determination module is configured to: analyzing the direction information and the amplitude information of the motion vector difference corresponding to the second original motion vector from the encoded bitstream of the current block; and determining the difference in motion vectors corresponding to the second original motion vector based on information about the direction and information about the amplitude of the difference in motion vectors corresponding to the second original motion vector. 14. The device according to item 13, in which, if it is determined that the direction information indicates the right direction and the amplitude information indicates that the amplitude is Ar, the motion vector difference is (Ar, 0); if it is determined that the direction information indicates the downward direction and the amplitude information indicates that the amplitude is Ad, the motion vector difference is (0, -Ad); if it is determined that the direction information indicates the left direction and the amplitude information indicates that the amplitude is Al, the motion vector difference is (-Al, 0); and if it is determined that the direction information indicates the upward direction and the amplitude information indicates that the amplitude is Au, the motion vector difference is (0, Au). 15. The device according to claim 11, wherein, when determining the first value of the prediction of the pixel position based on the first prediction mode of the current block and determining the second value of the prediction of the pixel position based on the second prediction mode of the current block, the third determining module is configured to: analyzing the first flag information from the encoded bitstream of the current block; and, if it is determined that the first flag information indicates that there is no overlap of the motion vector difference with the first source motion vector, selecting one possible motion information from the motion compensation candidate list as the first target motion information of the current block; and determining a first pixel position prediction value based on the first target motion information; and analyzing the second flag information from the encoded bitstream of the current block; and, if it is determined that the second flag information indicates that there is no overlap of the motion vector difference with the second source motion vector, selecting other possible motion information from the list of candidates for motion compensation as the second target motion information of the current block; and determining a second pixel position prediction value based on the second target motion information. 16. The device according to claim 11, wherein the second determining module is also configured to: configuring reference weight values for surrounding positions outside the current block, wherein the reference weight values of the surrounding positions outside the current block are configured based on the weight configuration parameters, and the weight configuration parameters include a weight conversion factor and a weight conversion start position, wherein the initial position of the weight transformation is determined by at least one of the following parameters: a weighted prediction angle of the current block, a weighted prediction position of the current block, or a size of the current block. 17. The device according to item 16, in which surrounding positions outside the current block contain surrounding positions one row above the current block or surrounding positions one column to the left of the current block, wherein the number of surrounding positions beyond the current block is determined based on the size of the current block and/or the weighted prediction angle of the current block, and the reference weight values of surrounding positions outside the current block monotonically increase or monotonically decrease. 18. The device according to item 16, in which the reference weight values of the surrounding positions outside the current block include one or more reference weight values of the target positions, one or more reference weight values of the first adjacent positions of the target positions and one or more reference weight values of the second adjacent positions of the target positions, wherein all of said one or more reference weight values of the first adjacent positions are second reference weight values, all of said one or more reference weight values of the second adjacent positions are third reference weight values, and the second reference weight values differ from the third reference weight values; the target positions include one reference weight value or at least two reference weight values; and, if it is determined that the target positions include at least two reference weight values, these at least two reference weight values of the target positions monotonically increase or monotonically decrease. 19. A device according to any one of paragraphs 11-18, in which the weighted prediction angle of the current block is the horizontal angle; or the weighted prediction angle of the current block is the vertical angle; or The absolute value of the slope of the weighted prediction angle of the current block is equal to the n-th power of 2, where n is an integer. 20. The device of any one of claims 11-18, wherein the surrounding positions outside the current block include one or more integer pixel positions, or one or more sub-pixel positions, or both one or more integer pixel positions and one or more sub-pixel positions; wherein when determining the target weight value of the pixel position based on the reference weight value associated with the surrounding corresponding position, the second determining module is configured to: if it is determined that the surrounding corresponding position is an integer pixel position, and the integer pixel position is set with a reference weight value, determining a target weight value of the pixel position based on the reference weight value of the integer pixel position; and, if it is determined that the surrounding corresponding position is a sub-pixel position, and the sub-pixel position is set with a reference weight value, determining a target weight value of the pixel position based on the reference weight value of the sub-pixel position. 21. A decoding device comprising: processor and machine-readable storage medium, wherein the machine-readable data carrier stores machine-readable instructions executable by the processor to implement the method according to any of paragraphs 1-10. 22. A machine-readable storage medium that stores machine-readable instructions executable by a processor to implement the method according to any one of paragraphs 1-10.

Images