WO2025190039A1 - 图像处理方法、装置、存储介质、电子设备及程序产品 - Google Patents

图像处理方法、装置、存储介质、电子设备及程序产品

Info

Publication number
WO2025190039A1
WO2025190039A1 PCT/CN2025/077524 CN2025077524W WO2025190039A1 WO 2025190039 A1 WO2025190039 A1 WO 2025190039A1 CN 2025077524 W CN2025077524 W CN 2025077524W WO 2025190039 A1 WO2025190039 A1 WO 2025190039A1
Authority
WO
WIPO (PCT)
Prior art keywords
image block
color component
image
area
block
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/CN2025/077524
Other languages
English (en)
French (fr)
Inventor
王力强
许晓中
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology (Shenzhen) Co Ltd
Original Assignee
Tencent Technology (Shenzhen) Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tencent Technology (Shenzhen) Co Ltd filed Critical Tencent Technology (Shenzhen) Co Ltd
Priority to KR1020267019671A priority Critical patent/KR20260113116A/ko
Publication of WO2025190039A1 publication Critical patent/WO2025190039A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/186—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being a colour or a chrominance component
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/103—Selection of coding mode or of prediction mode
    • H04N19/105—Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/124—Quantisation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/593—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving spatial prediction techniques

Definitions

  • the original audio and video data must be encoded at the data transmitter to form a compressed data stream. After the data stream is transmitted to the data receiver, it is decoded to restore the predicted and reconstructed audio and video data.
  • the present application provides an image processing method, an image processing device, a computer-readable storage medium, an electronic device, and a computer program product, the purpose of which is to improve the efficiency and accuracy of video encoding and decoding.
  • an embodiment of the present application provides an image processing method, performed by an electronic device, the method comprising:
  • the reconstructed value of the first color component of the target area is mapped according to the mapping relationship between the first color component and the second color component to obtain a predicted value of the second color component of the second image block.
  • an image processing device comprising:
  • a determination module configured to determine, based on a position of the second image block in the current video frame, a coding region associated with the position in the at least one first image block as a target region;
  • a prediction module configured to predict and obtain a reconstructed value of the first color component of the target area according to a preset prediction mode
  • a mapping module is configured to map the reconstructed value of the first color component of the target area according to a mapping relationship between the first color component and the second color component in response to a preset prediction condition being met, so as to obtain a predicted value of the second color component of the second image block.
  • an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the image processing method in the above technical solution when executed by a processor.
  • an embodiment of the present application provides a computer program product, including a computer program, which is stored in a computer-readable storage medium; when a processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device performs the image processing method in the above technical solution.
  • FIG1 schematically shows a diagram of an exemplary system architecture to which the technical solutions of the embodiments of the present application can be applied.
  • FIG2 schematically shows how a video encoding device and a video decoding device are placed in a streaming environment.
  • FIG4 shows a flowchart of an image processing method in one embodiment of the present application.
  • FIG5 is a schematic diagram showing a process of sampling color components of a current block in one embodiment of the present application.
  • FIG6 schematically shows the division results of various block division modes.
  • FIG. 7 shows image blocks obtained by performing image segmentation on luminance components and chrominance components using different segmentation tree structures in an application scenario.
  • FIG8 shows a flowchart of fitting a prediction model for representing a mapping relationship between color components in one embodiment of the present application.
  • FIG9 is a schematic diagram showing the relationship between a designated pixel point and neighboring pixel points in one embodiment of the present application.
  • FIG11 shows a schematic diagram of a region template in which some sub-regions are combined into a reference region in one embodiment of the present application.
  • FIG12 is a schematic diagram showing a region template of a reference region selected for a current block in one embodiment of the present application.
  • FIG. 13 is a schematic diagram showing a method of selecting a sampling position based on the position coordinates of a pixel point in one embodiment of the present application.
  • FIG14 is a schematic diagram showing a method of selecting sampling positions based on a round-trip scanning method in one embodiment of the present application.
  • FIG15 shows a schematic diagram of selecting sampling positions based on the ZigZag scanning method in one embodiment of the present application.
  • FIG16 shows a schematic diagram of a sampling window for downsampling the first color component in one embodiment of the present application.
  • FIG18 schematically shows a structural block diagram of an electronic device suitable for implementing an embodiment of the present application.
  • module refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof.
  • a processor or multiple processors or memories
  • each module or unit can be part of an overall module or unit that includes the function of the module or unit.
  • Video coding refers to the processing of a sequence of pictures to form a video or video sequence.
  • picture In the field of video coding, the terms “picture”, “frame” or “image” can be used as synonyms.
  • the video coding (coding) used in the embodiments of the present application represents video encoding (encoding) or video decoding (decoding).
  • video encoding (encoding) is performed on the source side, including processing (for example, compressing) the original video image to reduce the amount of data required to represent the video image, thereby more efficiently storing and/or transmitting.
  • Video decoding is performed on the destination side, and performs inverse processing relative to the encoder to reconstruct the video image.
  • the "coding" of video frames involved in the embodiments of the present application can also be understood as “encoding” or "decoding" a video image sequence.
  • the combination of the encoding part and the decoding part is also called encoding and decoding.
  • each image in a video image sequence is divided into a set of non-overlapping blocks and encoded at the block level.
  • the encoder side encodes the video, that is, at the block (also called image block or video block) level, including generating a prediction block through spatial (intra-image) prediction and temporal (inter-image) prediction, subtracting the prediction block from the current block (the currently processed block or the block to be processed) to obtain a residual block, and transforming and quantizing the residual block in the transform domain to reduce the amount of data to be transmitted (or to be compressed).
  • the decoder side applies the inverse processing part relative to the encoder to the encoded or compressed block to reconstruct the current block.
  • the encoder replicates the loop process on the decoder side, so that the encoder and decoder generate the same predicted (eg, intra-frame predicted and inter-frame predicted) blocks and/or reconstructed blocks for encoding subsequent blocks.
  • predicted eg, intra-frame predicted and inter-frame predicted
  • block refers to a portion of an image or frame.
  • the current block refers to the block currently being processed.
  • encoding it refers to the block currently being encoded
  • decoding it refers to the block currently being decoded.
  • FIG1 schematically shows a diagram of an exemplary system architecture to which the technical solutions of the embodiments of the present application can be applied.
  • a system architecture 100 includes a plurality of terminal devices that can communicate with each other via, for example, a network 150.
  • the system architecture 100 may include a first terminal device 110 and a second terminal device 120 interconnected via the network 150.
  • the first terminal device 110 and the second terminal device 120 perform unidirectional data transmission.
  • the first terminal device 110 may encode video data (e.g., a video picture stream captured by the terminal device 110) for transmission to the second terminal device 120 via the network 150.
  • the encoded video data is transmitted in the form of at least one encoded video stream.
  • the second terminal device 120 may receive the encoded video data from the network 150, decode the encoded video data to restore the video data, and display the video picture based on the restored video data.
  • the system architecture 100 may include a third terminal device 130 and a fourth terminal device 140, which can perform bidirectional transmission of encoded video data.
  • the bidirectional transmission can occur during a video conference.
  • each of the third terminal device 130 and the fourth terminal device 140 can encode video data (e.g., a video image stream captured by the terminal device) for transmission to the other of the third terminal device 130 and the fourth terminal device 140 via the network 150.
  • Each of the third terminal device 130 and the fourth terminal device 140 can also receive the encoded video data transmitted by the other of the third terminal device 130 and the fourth terminal device 140, decode the encoded video data to recover the video data, and display the video image on an accessible display device based on the recovered video data.
  • the first terminal device 110, the second terminal device 120, the third terminal device 130, and the fourth terminal device 140 may be servers, personal computers, and smartphones, but the present application is not limited thereto.
  • the embodiments disclosed herein are applicable to laptop computers, tablet computers, media players, and/or dedicated video conferencing equipment.
  • Network 150 represents any number of networks for transmitting encoded video data between the first terminal device 110, the second terminal device 120, the third terminal device 130, and the fourth terminal device 140, including, for example, wired and/or wireless communication networks.
  • the communication network 150 may exchange data using circuit-switched and/or packet-switched channels.
  • the network may include a telecommunications network, a local area network, a wide area network, and/or the Internet.
  • FIG2 schematically illustrates the placement of a video encoding device and a video decoding device in a streaming environment.
  • the subject matter disclosed herein is equally applicable to other video-enabled applications, including, for example, video conferencing, digital television, and storing compressed video on digital media such as CDs, DVDs, and memory sticks.
  • the streaming system may include an acquisition subsystem 213, which may include a video source 201, such as a digital camera.
  • the video source creates an uncompressed video picture stream 202.
  • the video picture stream 202 includes samples captured by the digital camera.
  • the video picture stream 202 is depicted as a thick line to emphasize the high data volume of the video picture stream.
  • the video picture stream 202 may be processed by an electronic device 220.
  • the electronic device 220 includes a video encoding device 203 coupled to the video source 201.
  • the video encoding device 203 may include hardware, software, or a combination of hardware and software to implement or embody various embodiments as described in more detail below.
  • the encoded video data 204 (or the encoded video code stream 204) is depicted as a thin line to emphasize the lower amount of data of the encoded video data 204 (or the encoded video code stream 204), which can be stored on the streaming server 205 for future use.
  • At least one streaming client subsystem such as the client subsystem 206 and the client subsystem 208 in FIG. 2 , can access the streaming server 205 to retrieve the copies 207 and 209 of the encoded video data 204.
  • the client subsystem 206 may include, for example, a video decoding device 210 in an electronic device 230.
  • the video decoding device 210 decodes an incoming copy of the encoded video data 207 and outputs a video picture stream 211, which may be presented on a display 212 (e.g., a screen) or another presentation device.
  • the encoded video data 204, video data 207, and video data 209 may be encoded according to some video encoding/compression standard.
  • the electronic device 220 and the electronic device 230 may include other components not shown in the figure.
  • the electronic device 220 may include a video decoding device
  • the electronic device 230 may also include a video encoding device.
  • HEVC High Efficiency Video Coding
  • VVC Very Video Coding
  • AVS Analog Video Coding Standard
  • CTU Coding Tree Unit
  • LCU Large Coding Unit
  • FIG3 schematically shows a basic flow chart of a video encoder, in which intra-frame prediction is taken as an example for explanation.
  • the original image signal s k [x,y] and the predicted image signal Perform difference operation to obtain the residual signal u k [x,y].
  • the quantized coefficient is obtained.
  • the quantized coefficient is entropy coded to obtain the encoded bit stream; on the other hand, the quantized coefficient is inversely quantized and inversely transformed to obtain the reconstructed residual signal u′ k [x,y].
  • the predicted image signal Superimposed with the reconstructed residual signal u'k [x,y] to generate the image signal
  • the image signal Input to the intra-frame mode decision module and the intra-frame prediction module for intra-frame prediction processing; on the other hand, the image signal Perform loop filtering and output the reconstructed image signal s'k [x,y].
  • the reconstructed image signal s'k [x,y] can be used as the reference image for the next frame for motion estimation and motion compensation prediction. Then, based on the motion compensation prediction result s'r [x+ mx ,y+ my ] and the intra-frame prediction result Get the predicted image signal of the next frame And continue to repeat the above process until the encoding is completed.
  • Predictive Coding includes intra-frame prediction and inter-frame prediction.
  • the original video signal is predicted by the selected reconstructed video signal to obtain a residual signal.
  • the encoder needs to decide which prediction mode (also known as encoding mode) to select for the current CU and inform the decoder.
  • intra-frame prediction Intra Prediction
  • Inter Prediction means that the predicted signal comes from another image that has been encoded and different from the current image (called a reference image).
  • Transform & Quantization After the residual signal undergoes transformations such as DFT (Discrete Fourier Transform) and DCT (Discrete Cosine Transform), it is converted to the transform domain, known as transform coefficients. The transform coefficients are then subjected to a lossy quantization operation, which loses some information, making the quantized signal more suitable for compression. In some video coding standards, more than one transform method may be available, so the encoder must select one for the current CU and inform the decoder. The level of quantization is determined by the quantization parameter (QP). A larger QP value means that coefficients with a larger value range will be quantized to the same output, resulting in greater distortion and a lower bitrate. Conversely, a smaller QP value means that coefficients with a smaller value range will be quantized to the same output, resulting in less distortion and a corresponding higher bitrate.
  • QP quantization parameter
  • Entropy Coding or Statistical Coding The quantized transform domain signal will be statistically compressed and encoded according to the frequency of occurrence of each value, and finally a binary (0 or 1) compressed code stream will be output. At the same time, other information generated by the encoding, such as the selected prediction mode, motion vector data, etc., also needs to be entropy encoded to reduce the bit rate.
  • Statistical coding is a lossless coding method that can effectively reduce the bit rate required to express the same signal. Common statistical coding methods include variable length coding (VLC) or context-based binary arithmetic coding (CABAC).
  • Bypass Coding Mode Instead of assigning a specific probability model to each binary bit, the input binary bit value is directly encoded using a simple bypass encoder, speeding up the entire encoding and decoding process.
  • different syntax elements are not completely independent, and the same syntax element itself has a certain degree of memory. Therefore, according to conditional entropy theory, conditional encoding using other encoded syntax elements can further improve encoding performance compared to independent encoding or memoryless encoding. This encoded symbol information used as a condition is called context.
  • the bins of syntax elements enter the context modeler sequentially.
  • the encoder assigns an appropriate probability model to each input bin based on the value of the previously encoded syntax element or bin. This process is called context modeling.
  • the context model corresponding to the syntax element can be located through ctxIdxInc (context index increment) and ctxIdxStart (context index Start).
  • the context model needs to be updated according to the bin value, which is the adaptive process in encoding.
  • Loop Filtering The changed and quantized signal will be reconstructed through inverse quantization, inverse transformation and prediction compensation operations to obtain a reconstructed image.
  • the reconstructed image has some information that is different from the original image due to the influence of quantization, that is, the reconstructed image will produce distortion. Therefore, the reconstructed image can be filtered, such as deblocking filter (DB), SAO (Sample Adaptive Offset) or ALF (Adaptive Loop Filter) and other filters, which can effectively reduce the degree of distortion caused by quantization. Since these filtered reconstructed images will be used as a reference for subsequent encoded images to predict future image signals, the above filtering operation is also called loop filtering, that is, filtering operation within the encoding loop.
  • DB deblocking filter
  • SAO Sample Adaptive Offset
  • ALF Adaptive Loop Filter
  • the decoding end after obtaining the compressed code stream (i.e., bitstream), entropy decoding is performed on each CU to obtain various mode information and quantization coefficients. The quantization coefficients are then dequantized and inversely transformed to obtain a residual signal.
  • the prediction signal corresponding to the CU can be obtained.
  • the residual signal is then added to the prediction signal to obtain the reconstructed signal.
  • the reconstructed signal then undergoes loop filtering and other operations to produce the final output signal.
  • FIG 4 shows a flowchart of an image processing method in an embodiment of the present application.
  • the image processing method can be executed by an electronic device, for example, by a terminal device or a server that sends or receives video encoding data.
  • the embodiment of the present application uses the method executed by a terminal device as an example to illustrate the method.
  • the terminal device can be, for example, the video decoding device 210 or the video encoding device 220 shown in Figure 2.
  • the image processing method in the embodiment of the present application includes the following steps S410 to S440 .
  • S410 Segment the current video frame to obtain at least one first image block and a second image block, wherein the first image block is a coding unit having a first color component, and the second image block is a coding unit having a second color component;
  • S420 Determine, according to the position of the second image block in the current video frame, a coding region associated with the position in at least one first image block as a target region;
  • S430 Predicting and obtaining a reconstruction value of the first color component of the target area according to a preset prediction mode
  • the prediction modes preset in S430 include intra-frame prediction mode, inter-frame prediction mode or intra-frame block copy prediction mode.
  • the predicted pixels used when using these prediction modes are obtained by searching in the reconstructed reference frame or the reconstructed area of the current frame or searching in its subpixel interpolated image.
  • the embodiment of the present application adopts a color component mapping prediction method to predict the second color component, thereby realizing cross-component prediction for chrominance, overcoming the problem of poor prediction accuracy caused by independent prediction of different color components in traditional coding schemes, enabling cross-component prediction to be normally used in the luminance and chrominance separation tree, expanding the scope of use of cross-component prediction, and improving the performance of the cross-component prediction method, thereby improving the encoding and decoding efficiency and accuracy of the video.
  • an image block may correspond to a color category.
  • color component sampling may be performed on the current block in a color space to form a luminance block and a chrominance block.
  • Color video images contain not only a luma component (Y) but also chroma components (U, V). Such images are also called YUV images.
  • YUV images When encoding YUV images, in addition to encoding the luma component, the chroma components also need to be encoded. Because the human eye is more sensitive to brightness and less so to color, to save storage space and improve coding efficiency, the luma component can be sampled at full resolution, while the chroma components do not need to be sampled at full resolution.
  • video sequences typically have various formats: 4:4:4 YUV images, 4:2:2 YUV images, 4:2:0 YUV images, and so on.
  • Figure 5 shows a schematic diagram of the process of sampling the color components of the current block in one embodiment of the present application.
  • the current block 501 is an image block of size 16 ⁇ 16.
  • different resolution formats can be selected for sampling.
  • the 4:4:4 format indicates that the chroma components are not downsampled.
  • the 4:4:4 format has the highest resolution for chroma components.
  • one Y component corresponds to a set of UV components.
  • the data in four adjacent pixels contains 4 Y, 4 U, and 4 V components.
  • the 4:2:2 format means that the chroma components are horizontally downsampled 2:1 relative to the luma components, with no vertical downsampling. For every two U or V samples, each row contains four Y samples. When sampling in the 4:2:2 format, every two Y components share a set of UV components, and the data in four adjacent pixels contains 4 Y, 2 U, and 2 V components.
  • the 4:2:0 format means that the chroma components are horizontally downsampled 2:1 relative to the luma components, and vertically downsampled 2:1.
  • the 4:2:0 format is the format with the lowest resolution for chroma components.
  • the chroma samples are only half of the luma samples in each row (i.e. horizontally), and only half of the luma samples in each column (i.e. vertically).
  • the U component and the Y component appear alternately. For example, if the three YUV components appear in a ratio of 4:2:0 in the first row, then in the second row, the ratio is 4:0:2; there are 4 Y, 1 U, and 1 V in 4 adjacent pixels.
  • the chrominance component of the image block is an M ⁇ N image block. For example, if the resolution of the image block is 720*480, then the resolution of the luminance component of the image block is 720*480, and the resolution of the chrominance component of the image block is 360*240.
  • the 4:2:0 format is used as an example.
  • a luminance block 502 and a chrominance block 503 are obtained.
  • the luminance block 502 is a 16 ⁇ 16 image block
  • the corresponding chrominance block 503 is an 8 ⁇ 8 image block.
  • the image area adjacent to the luminance block 502 and the chrominance block 503 is shown at 504.
  • the current video frame is segmented to obtain multiple image regions, and each image region or a combination of multiple image regions can be regarded as an image block.
  • the image blocks of the first color component and the image blocks of the second color component can be segmented according to the same segmentation tree structure (i.e., a single segmentation tree structure) to form image blocks in a merged mode, or can be segmented according to different segmentation tree structures to form image blocks in a separated mode.
  • the first color component can be any one of the luminance component Y, the chrominance component U, and the chrominance component V
  • the second color component can be another one of the luminance component Y, the chrominance component U, and the chrominance component V.
  • the segmentation tree structure in AVS3 generally consists of a quadtree, a binary tree, and an extended quadtree.
  • the luminance blocks and chrominance blocks follow the same block partitioning structure, and the luminance blocks and chrominance blocks are one-to-one corresponding, that is, one chrominance block corresponds to only one luminance block.
  • the luminance and chrominance blocks are divided into blocks independently, and the luminance blocks and chrominance blocks are no longer one-to-one corresponding. That is, the luminance area corresponding to a chrominance block may be in different luminance blocks.
  • a frame of image in a video signal can be divided into non-overlapping coding tree units (CTUs).
  • the block size of a CTU can be set to 64 ⁇ 64, 128 ⁇ 128, etc.
  • a CTU with a block size of 128 ⁇ 128 can be a matrix pixel array containing 128 columns, each column containing 128 pixels, and each pixel containing a luminance component and/or a chrominance component.
  • the CTU can be divided into a finer granularity to obtain a group of coding units (CUs), each of which can contain at least one CU.
  • CUs group of coding units
  • the coding tree unit can be divided into one of the following division methods: QT (Quaternary Tree), BT (Binary Tree), and TT (Ternary Tree), or a combination of multiple division methods.
  • BT can further include HBT (Horizontal Binary Tree) and VBT (Vertical Binary Tree)
  • TT can further include HTT (Horizontal Ternary Tree) and VTT (Vertical Ternary Tree).
  • the CTU can be used as the root node and divided into several nodes. Each node corresponds to an image region. If a node is no longer divided, it is called a leaf node, and the image region corresponding to it forms a CU. If the node is further divided, one or a combination of the above division methods can be used to divide the node into multiple sub-regions, each sub-region corresponding to a sub-node. It is then necessary to determine whether these sub-nodes will be further divided.
  • the division level of the child node can be the division level of the parent node plus 1.
  • the encoder sets a minimum block size for the CU. During the division process, if the block size of a node is equal to the minimum block size, the node will no longer be divided by default. For the sake of convenience, the "image area corresponding to the node" will be referred to as "node" below.
  • the QT division mode can be used to divide the node into 4 child nodes. According to the QT division mode, the node can be divided into four child nodes of the same block size. The width and height of each child node are the same, and the width is half of the width of the node before the division, and the height is half of the height of the node before the division.
  • Figure 6 schematically illustrates the results of various block partitioning modes. As shown in Figure 6, for a node 610 with a block size of 64 ⁇ 64, if the node 610 is not partitioned further, the node 610 is directly determined to be a CU with a block size of 64 ⁇ 64. If the node 610 is partitioned further, it can be partitioned into four nodes 621-624 with a block size of 32 ⁇ 32 according to the QT partitioning mode. For one of these four nodes 621-624 with a block size of 32 ⁇ 32, if the node is partitioned further, four nodes with a block size of 16 ⁇ 16 are generated.
  • the BT partitioning mode can be used to divide the node into two child nodes.
  • the BT partitioning mode includes two types: HBT partitioning mode and VBT partitioning mode.
  • the HBT partitioning mode is to divide the node into two upper and lower child nodes of the same block size, and the width and height of each child node are the same, and the width is equal to the width of the node before the division, and the height is half of the height of the node before the division.
  • VBT is to divide the node into two left and right child nodes of the same size, and the width and height of each child node are the same, and the width is half of the width of the node before the division, and the height is equal to the height of the node before the division.
  • the node 610 can be divided into two nodes 631 and 632 with a block size of 64 ⁇ 32, or if the VBT partitioning mode is used, the node 610 can be divided into two nodes 641 and 642 with a block size of 32 ⁇ 64.
  • the TT partitioning mode can be used to divide the node into three child nodes.
  • the TT partitioning mode includes two types: HTT partitioning mode and VTT partitioning mode.
  • the HTT partitioning mode is to divide the node into three child nodes: upper, middle and lower, as shown in 651-653 in Figure 6, wherein the width of the upper and lower child nodes 651 and 653 is equal to the width of the node 610 before the partition, and the height is one-quarter of the height of the node 610 before the partition, and the width of the middle node 652 is equal to the width of the node before the partition, and the height is half of the height of the node before the partition.
  • VTT is to divide the node into three child nodes: left, middle and right, as shown in 661-663 in Figure 6, wherein the width of the left and right child nodes 661 and 663 is one-quarter of the width of the node 610 before the partition, and the height is equal to the height of the node 610 before the partition, and the width of the middle node 662 is half of the width of the node before the partition, and the height is equal to the width of the node before the partition.
  • the current video frame can be segmented based on the single segmentation tree structure to obtain at least one first image block and a second image block.
  • the entire area of a first image block corresponding to the second image block in at least one first image block can be determined as the target area.
  • the entire area of the first image block is determined as the target area.
  • the current video frame when the first image block and the second image block have different segmentation tree structures, can be segmented based on the first segmentation tree structure to obtain at least one first image block; and the current video frame can be segmented based on the second segmentation tree structure to obtain the second image block.
  • a partial area or all areas of the at least one first image block corresponding to the second image block can be determined as the target area. That is, in at least one first image block, there is not only one first image block corresponding to the second image block, then a partial area or all areas of the at least one first image block corresponding to the second image block can be determined as the target area.
  • the image segmentation scheme can support independent segmentation tree structures for the luminance block and chrominance block in a CTU.
  • the luminance block and chrominance block in a CTU share the same segmentation tree structure, that is, the luminance-chrominance merge mode.
  • the luminance block and chrominance block in a CTU can have different segmentation tree structures, that is, the luminance-chrominance separation mode.
  • the luminance block can be segmented into luminance CUs (which may be smaller in size) based on one coding tree structure, while the chrominance block can be segmented into chrominance CUs based on another coding tree structure.
  • the CU in the I frame can include a CU for the luminance component or a CU for two chrominance components, unless the video is monochrome;
  • the CU in the P frame or B frame can always include CUs for all three color components, that is, a CU for the luminance component and two CUs for the chrominance components.
  • a semi-decoupled tree (SDT) image segmentation scheme also known as a semi-separate tree (SST) or flexible block segmentation for chroma components
  • SDT semi-decoupled tree
  • the luma block and chroma block in a super block (SB) may have the same or different block segmentation schemes, depending on the block size of the luma block or the luma tree depth. For example, when the block size of the luma block is greater than a threshold T1 or the coding tree split depth of the luma block is less than or equal to a threshold T2, the chroma block associated with the luma block may use the same coding tree structure as the luma block.
  • Figure 7 shows image blocks obtained by segmenting the luminance and chrominance components using different segmentation tree structures in an application scenario.
  • the numbers 1, 2, and 3 indicate the order in which the segmentation is performed.
  • T2 is set to 1, meaning that when the luminance split depth of the luminance block is greater than 1, the associated chrominance block uses a different segmentation data structure.
  • 703 corresponds to 701 and 702 in the luminance block. Therefore, part or all of 701 and 702 can be identified as the target area.
  • luma and chroma blocks can share a partial tree structure starting from the root node of the superblock.
  • SDP semi-decoupled partitioning
  • Intra-frame prediction uses the reconstructed values of neighboring blocks. This process applies a pre-set weighting to the encoded adjacent pixels to obtain a better estimate of the current pixel block.
  • One-dimensional prediction uses the correlation between adjacent pixels in the same row. Because the next pixel value is close to the previous pixel value, the previous pixel value is used as the predicted value for the current pixel.
  • Two-dimensional prediction uses not only the neighboring pixels in the current row but also the pixels in the previous row. By assigning corresponding weights to the pixel values in different rows, a predicted value is obtained.
  • Inter-frame prediction uses the correlation between previous and next frames in a video sequence to predict the current frame.
  • adjacent frames have a high degree of pixel similarity, so information from the previous or next frame can be used to predict the pixel values of the current frame.
  • Inter-frame prediction is achieved through processes such as motion estimation, motion compensation, and residual coding.
  • Motion estimation compares the differences between the current frame and the previous or next frame to find the optimal motion vector that describes the motion relationship between the two frames. Common motion estimation algorithms include full search and block matching.
  • Motion compensation uses the found motion vector to modify the previous or next frame to predict the pixels of the current frame. Motion compensation can be achieved by applying the motion vector to a reference frame for pixel replication or interpolation.
  • Residual coding encodes the difference between the pixel values of the current frame and the pixel values of the predicted frame to generate residual data. This residual data represents pixel information in the current frame that cannot be obtained through motion prediction and requires additional encoding and transmission.
  • a prediction model for representing the mapping relationship between the first color component and the second color component can be fitted based on the current block and at least one image area in the reference area, and the reference area includes at least one decoded image area that is the nearest neighbor or the second nearest neighbor to the current block.
  • the second color component is another color component different from the first color component in the color space.
  • the second color component may be the chrominance component U or the chrominance component V.
  • Figure 8 shows a flowchart of fitting a prediction model for representing the mapping relationship between color components in one embodiment of the present application.
  • the method for obtaining the prediction model may include the following steps S810 to S830.
  • the input item of the initial prediction model also includes the first color component of at least one neighborhood pixel point, and the neighborhood pixel point is the pixel point that is the nearest neighbor or the second nearest neighbor of the designated pixel point.
  • FIG9 is a schematic diagram showing the relationship between a designated pixel point and neighboring pixel points in one embodiment of the present application.
  • C represents a designated pixel point
  • the designated pixel points in the first color component a and the second color component b may be pixel sample points at associated positions or the same position.
  • the neighborhood pixels of the designated pixel point C may include multiple nearest neighboring pixels, such as the multiple pixels N, S, W, and E located above, below, to the left, and to the right of the designated pixel point C as shown in FIG9 .
  • the neighborhood pixels of the designated pixel point C may include multiple next-nearest neighboring pixels, such as the multiple pixels NW, NE, SW and SE located at the upper left, upper right, lower left and lower right positions of the designated pixel point C as shown in FIG9 .
  • Each such connection between the first color component a and the second color component b is called a cross-component matching pair, generating a mapping equation. Multiple such mapping equations can be used to solve the model's weighted parameters.
  • the location of pixel C in Figure 9 is only an example; a wider range of pixel sample points can also be selected.
  • the initial prediction model includes at least one combination item with independent weighting parameters, and the combination item takes the first color components of at least two pixels as input items.
  • the initial prediction model in the embodiment of the present application can be composed of at least one monomial, wherein each monomial can have an independent weighting parameter.
  • each monomial can have an independent weighting parameter.
  • the monomial is called a combination item.
  • the initial prediction model may include at least two combinations of different orders, where the order is the highest power of the input items in the combination.
  • the power operation can introduce nonlinear factors to improve the model's fit to the color component mapping relationship.
  • the initial prediction model can be obtained by weighted operation of at least one of the following monomials: x, mx ⁇ ny, xy, xk , (mx ⁇ ny) k , (mx ⁇ ny)(pz ⁇ qf), (mx ⁇ ny)x, B.
  • m, n, p, and q represent weighting coefficients used to perform fixed weighting on the input items
  • x, y, z, and f represent the input items, i.e., the first color component of a specified pixel or a neighboring pixel
  • B represents a constant bias term
  • k represents the order of the power operation on the input item, where k is an integer greater than 1.
  • the initial prediction model may include at least one of the following multiple candidate models:
  • p 0 , p 1 , p 2 , p 3 , p 3 , p 4 , p 5 , p 6 , and p 7 represent weights.
  • S820 Acquire training sample pairs, where the training sample pairs include one or more of the following sample pairs:
  • the reconstructed value of the first color component and the reconstructed value of the second color component of the sample pixel point located in the reference area, the reference area includes at least one image area adjacent to the second image block.
  • a mapping relationship between the first color component and the second color component can be fitted based on the predicted values of the color components of the second image block and the corresponding target area.
  • the training sample pair used for fitting the mapping relationship can include (the predicted value of the first color component of the sample pixel point located in the target area, and the predicted value of the second color component of the sample pixel point located in the second image block).
  • the predicted value of the first color component of the sample pixel point located in the target area and the predicted value of the second color component of the sample pixel point located in the second image block are predicted according to a preset prediction mode, and the preset prediction mode includes an intra-frame prediction mode, an inter-frame prediction mode, or an intra-block copy prediction mode.
  • a mapping relationship between the first color component and the second color component can be fitted based on the reconstructed values of the color components of the reference area.
  • the training sample pair used for fitting the mapping relationship can include: (the reconstructed value of the first color component of a sample pixel located in the reference area, and the reconstructed value of the second color component of the sample pixel).
  • the mapping relationship between the first color component and the second color component can be fitted based on the predicted values of the color components of the second image block and the target area itself and the reconstructed values of the color components of the reference area.
  • the training sample pairs used to fit the mapping relationship can include the above two types, namely: (the predicted value of the first color component of the sample pixel point located in the target area, the predicted value of the second color component of the sample pixel point located in the second image block), and (the reconstructed value of the first color component of the sample pixel point located in the reference area, the reconstructed value of the second color component of the sample pixel point).
  • the reference area corresponding to the current block (unless otherwise specified, the current block described below refers to the second image block) is composed of at least one nearest neighbor area or second nearest neighbor area, the nearest neighbor area includes an image area with a specified image size located above or to the left of the current block, and the second nearest neighbor area includes an image area with a specified image size located above, below, or above the current block.
  • FIG10 shows a schematic diagram of the distribution of reference areas corresponding to the current block in one embodiment of the present application.
  • the reference region corresponding to the current block 1001 may be composed of multiple sub-regions 1002, wherein each sub-region 1002 may be the nearest neighbor region or the next nearest neighbor region of the current block.
  • the nearest neighbor region may, for example, include the image region B located above the current block 1001 or the image region D located to the left of the current block 1001, and the next nearest neighbor region may, for example, include the image region A located to the upper left of the current block 1001, the image region E located to the lower left of the current block 1001, or the image region C located to the upper right of the current block 1001.
  • the reference area of the current block 1001 can be formed by combining at least one of the image areas A-E.
  • the image sizes of the sub-regions constituting the reference region are specified as follows: the next-nearest neighboring region to the upper right of the second image block has the same image size as the second image block in the horizontal direction, and the next-nearest neighboring region to the upper right of the second image block has a specified image size in the vertical direction;
  • next nearest neighboring area at the lower left of the second image block has the same image size as the second image block in the vertical direction, and the next nearest neighboring area at the upper right of the second image block has a specified image size in the horizontal direction;
  • the nearest neighbor region D on the left side of the current block 1001 has the same image size as the current block 1001 in the vertical direction, and has a specified image size in the horizontal direction.
  • the next neighboring region E below the left of the current block 1001 has the same image size as the current block 1001 in the vertical direction, and the next neighboring region below the left of the current block 1001 has a specified image size in the horizontal direction.
  • the next neighboring area A at the upper left of the current block 1001 has a specified image size in both the horizontal and vertical directions.
  • the width and/or height of the specified image size may be a preset value greater than or equal to one.
  • the width of the specified image size may be set to 6.
  • color component prediction may be performed using multiple layers of neighboring pixels, thereby improving the accuracy of color component prediction.
  • the sub-regions of the reference region constituting the current block may have the same specified image size or different specified image sizes.
  • the vertical size of image region C may be the same as or different from the horizontal size of image region E.
  • pixels that have been decoded and reconstructed or pixels that are allowed to be used can be selected from the sub-region as reference pixels for color component prediction of the current block. For example, when some pixels in image region C have been decoded and reconstructed, while other pixels have not been decoded and reconstructed, pixels that have been decoded and reconstructed can be selected from image region C as reference pixels for color component prediction of the current block.
  • the sub-neighboring region when the decoded and reconstructed pixels in a sub-region do not meet the aforementioned size requirements, the sub-neighboring region can be configured as unavailable.
  • image region C when the lower right corner of image region C is not reconstructed or exceeds the image boundary, image region C can be configured as unavailable; for another example, when the lower right corner of image region E is not reconstructed or exceeds the image boundary, image region E can be configured as unavailable.
  • all of the sub-regions shown in FIG. 8 may be combined to form a reference region of the current block, or a portion of the sub-regions may be combined to form a reference region of the current block.
  • Figure 11 shows a schematic diagram of a region template that combines some subregions into a reference region in one embodiment of the present application. As shown in Figure 11, based on the combination of different subregions, ten exemplary candidate region templates can be formed. In this embodiment of the present application, one or more of these candidate region templates can be specified for the current block.
  • Figure 12 shows a schematic diagram of a region template of a reference region selected for a current block in one embodiment of the present application.
  • the reference region corresponding to the current block includes at least one of a full region combination, a left region combination, or an upper region combination.
  • the full region combination includes the nearest neighbor regions located on the left and above the current block and the next nearest neighbor regions located on the upper left, lower left and upper right of the current block.
  • the left region combination includes the nearest neighbor region located on the left side of the current block and the next nearest neighbor region located on the lower left side of the current block.
  • the upper region combination includes the nearest neighbor region located above the current block and the next nearest neighbor region located above and to the right of the current block.
  • an indicator field can be used in the video bitstream to identify the region template of the reference region used by the current block. For example, when the indicator field value is 1, it indicates that the reference region selected by the current block is the full region combination shown in Figure 12; when the indicator field value is 01, it indicates that the reference region selected by the current block is the left region combination shown in Figure 12; when the indicator field value is 00, it indicates that the reference region selected by the current block is the upper region combination shown in Figure 12.
  • availability information of at least one sub-area constituting the reference area can be obtained, and then the area range of the reference area can be adjusted based on the availability information of at least one sub-area.
  • adjusting the area range of the reference area based on the availability information of the at least one sub-area may further include: removing the sub-areas in an unavailable state from the reference area; when all sub-areas in the reference area are in an unavailable state, configuring the reference area to an unavailable state.
  • the indicator field corresponding to the current block has a value of 1, indicating that the reference area used for color component prediction of the current block is the full area combination including the five sub-areas A-E shown in Figure 12.
  • availability information of each sub-region in the reference region may be obtained, and the region range of the reference region may be adjusted according to the availability information.
  • sub-regions A, B, and C are image regions of other image blocks located above the current block. If the image blocks where sub-regions A, B, and C are located have not yet completed encoding and reconstruction, sub-regions A, B, and C are in an unavailable state. At this time, the area range of the reference area can be adjusted from A+B+C+D+E to D+E.
  • sub-regions D and E are image regions of other image blocks located to the left of the current block. If the image blocks containing sub-regions D and E have not yet been coded and reconstructed, sub-regions D and E are also unavailable. In this case, all five sub-regions A-E are unavailable, so the entire reference region based on this full region combination can be configured as unavailable.
  • the sample pixels used to train the prediction model may be all or a portion of the pixels selected from at least one of the second image block and the reference area.
  • pixel sampling may be performed within at least one of the second image block and the reference area according to a preset sampling rule.
  • selecting a training sample pair from at least one image region of the second image block and the reference region may include:
  • the sample pixel points located in the second image block refer to all pixel points in the second image block
  • the sample pixel points located in the reference area refer to all pixel points in at least one image area in the reference area
  • the sample pixel points located in the second image block refer to the partial pixel points at the specified sampling position
  • the sample pixel points located in the reference area refer to the partial pixel points at the specified sampling position in at least one image area in the reference area.
  • specifying the sampling location includes satisfying at least one of the following conditions.
  • the first sampling location specifies the coordinates of the sampling location and satisfies the preset coordinate value conditions.
  • the coordinate value condition includes: at least one of the horizontal position coordinate or the vertical position coordinate of the pixel point is an even number; or, at least one of the horizontal position coordinate or the vertical position coordinate of the pixel point is an odd number.
  • FIG. 13 is a schematic diagram showing a method of selecting a sampling position based on the position coordinates of a pixel point in one embodiment of the present application.
  • a coordinate system is established with the upper left corner as the coordinate origin (0, 0), the horizontal position coordinate x represents the sequential position of the pixel points arranged from left to right in the horizontal direction, and the vertical position coordinate y represents the sequential position of the pixel points arranged from top to bottom in the horizontal direction.
  • even positions, odd positions or all positions can be selected as designated sampling positions in the horizontal direction, and even positions, odd positions or all positions can be selected as designated sampling positions in the vertical direction.
  • the second sampling position select the specified sampling position along the preset pixel scanning direction.
  • pixels in the reference area can be scanned using any scanning method, such as a round-trip scan or a ZigZag scan, so that designated sampling positions are selected along a preset pixel scanning direction in accordance with a sequence and a preset sampling rule.
  • the preset sampling rule may be, for example, interval sampling, where a designated sampling position is selected after at least one scanned pixel interval.
  • FIG14 is a schematic diagram showing a method of selecting sampling positions based on a round-trip scanning method in one embodiment of the present application.
  • FIG15 shows a schematic diagram of selecting sampling positions based on the ZigZag scanning method in one embodiment of the present application.
  • the third sampling position the reconstructed value of the first color component of the pixel point at the specified sampling position is within a preset value range.
  • the embodiment of the present application can select the position of the pixel point whose brightness value is greater than or less than a certain threshold as the designated sampling position.
  • a prediction model can be fitted by selecting sample pixel points in the reference area whose brightness values are greater than a set threshold, and another prediction model can be fitted by selecting sample pixel points whose brightness values are less than or equal to the set threshold; when generating a predicted image, the corresponding first prediction model can be used for pixel points whose brightness values are greater than the set threshold, and the corresponding second prediction model can be used for pixel points whose brightness values are less than or equal to the set threshold.
  • S830 Perform parameter fitting on the initial prediction model according to the training sample pairs to obtain the prediction model.
  • the prediction model can fit the model parameters through online training or offline training.
  • A is a matrix with M rows and N columns, M and N represent the number of equations established according to the initial prediction model and the number of model parameters in the initial prediction model, respectively, and x represents a parameter vector composed of the model parameters in the initial prediction model as elements.
  • the LDL decomposition method or the Gaussian elimination method can be selected to solve the above equation.
  • the solution steps may include:
  • the decomposition processes of ATA in the two chrominance component prediction processes can be merged at this time, thereby reducing the computational complexity and improving the encoding and decoding efficiency.
  • the reconstructed value of the first color component of the target area can be used as an input item and input into the target prediction model, and the target prediction model outputs the predicted value of the second color component of the second image block.
  • the color component prediction scheme provided in the above embodiment can be executed on a portion of the image blocks in the current video frame, while the color component prediction scheme provided in the above embodiment can be prohibited on another portion of the image blocks.
  • the prediction conditions corresponding to the image blocks can be explicitly or implicitly identified, and when the prediction conditions are met, the color component prediction method for predicting the second color component based on the first color component is executed.
  • the prediction condition when the first image block and the second image block have the same segmentation tree structure, the prediction condition includes: the number of non-zero coefficients of the first color component in the target area is greater than a preset threshold.
  • the non-zero coefficient of the first color component refers to the quantization coefficient of the first color component in the first image block whose value is not zero.
  • the quantity threshold can be a preset default value, or can also be parsed or implicitly derived through the video code stream.
  • the luminance component when the chroma block and the luminance block have the same partition tree structure, that is, in the merged tree mode, if the number of non-zero coefficients in the luminance block corresponding to the chroma block is greater than or equal to the specified threshold, the luminance component is allowed to be used to predict the chroma component, otherwise the luminance component is prohibited from being used to predict the chroma component.
  • the prediction conditions include one or more of the following conditions: the number of non-zero coefficients of the first color component in the target area is greater than a preset number threshold; the sum of the numbers of non-zero coefficients of the first color components of multiple first image blocks corresponding to the target area is greater than a preset number threshold.
  • the chroma block and the luminance block have different segmentation tree structures, that is, in the separate tree mode, if the number of non-zero coefficients in the luminance area corresponding to the chroma block (that is, part of the area or all of the area in at least one luminance block corresponding to the chroma block) is greater than or equal to the specified threshold, the use of the luminance component to predict the chroma component is allowed, otherwise the use of the luminance component to predict the chroma component is prohibited.
  • the luminance component when the chroma block and the luminance block have different segmentation tree structures, that is, in the separate tree mode, if the sum of the number of non-zero coefficients in all luminance blocks where the luminance area corresponding to the chroma block is located is greater than or equal to the specified threshold, the luminance component is allowed to be used to predict the chroma component, otherwise the luminance component is prohibited from being used to predict the chroma component.
  • the prediction conditions include one or more of the following conditions: the first image block and the second image block have the same segmentation tree structure, and the number of non-zero coefficients of the first color component in the target area is greater than a preset number threshold; or, the first image block and the second image block have different segmentation tree structures.
  • cross-component prediction can be used regardless of the number of non-zero coefficients in the corresponding luma block.
  • the prediction condition may further include at least one of the following five conditions:
  • the first prediction condition using a preset prediction mode to predict the second image block, the prediction mode includes one or more of an intra-frame prediction mode, an inter-frame prediction mode and an intra-block copy prediction mode.
  • the second prediction condition the index corresponding to the second image block has a specified index value, and the index is used to indicate whether to execute: mapping the reconstructed value of the first color component to obtain the predicted value of the second color component of the second image block.
  • an index can be decoded in the sequence header/image header/slice header/maximum coding block, which is used to indicate whether all image blocks in the current sequence/image/slice/maximum coding block perform a cross-component prediction method for predicting the second color component based on the first color component.
  • the third prediction condition the image features of the target area meet the preset first feature condition.
  • the first characteristic condition includes: a reconstruction parameter of a first color component in the target area is within a preset value range.
  • the preset numerical range includes greater than zero or less than zero.
  • the fourth color component prediction condition the image feature of the second image block satisfies a preset second feature condition.
  • the second characteristic condition may include: the image size of the second image block is within a preset size range; or the position of the second image block in the current video frame is within a specified area range.
  • whether the second image block is to be used for a cross-component prediction method for predicting the second color component based on the first color component may be indicated based on whether the position of the second image block meets a preset range limit. For example, if the second image block is located in the upper left corner of the current video frame, the cross-component prediction method for predicting the second color component based on the first color component is not used.
  • the fifth color component prediction condition the image characteristics of the reference area corresponding to the second image block meet the preset third characteristic condition, and the reference area includes at least one image area adjacent to the second image block.
  • the third characteristic condition includes: the area of the reference region corresponding to the second image block is greater than a specified area threshold; or the number of pixels at a specified sampling position in the reference region corresponding to the second image block is greater than a preset number threshold.
  • the preset number threshold may be the number of model parameters in the prediction model.
  • the cross-component prediction method of predicting the second color component based on the first color component is performed on the second image block only when the area of the reference region is greater than a specified area threshold.
  • the cross-component prediction method of predicting the second color component based on the first color component is performed on the second image block only when the number of pixels at a specified sampling position in the reference area (i.e., the number of pixels included in a matching pair of the first color component and the second color component) is greater than the number of model parameters in the prediction model.
  • the first color component before mapping the reconstructed value of the first color component of the target area according to the mapping relationship, the first color component may be downsampled to obtain a first color component that can form a matching pair with the second color component.
  • the luma and chroma blocks obtained after color component sampling according to the 420 sampling format have different block sizes.
  • the resolution of the luma component is twice that of the chroma components in both the horizontal and vertical directions. In this case, to form a matching pair of luma and chroma components, the luma component can be downsampled.
  • a method for downsampling a first color component may include: determining a sampling window having a specified window size based on pixel locations of the second color component of a second image block; downsampling a reconstructed value of the first color component of a target area within the sampling window to obtain a reconstructed value of the first color component that matches the pixel location.
  • the matching reconstructed value of the first color component may be mapped to obtain a predicted value of the second color component of the second image block.
  • the sampling window determined according to the pixel position of the second color component is used to cover the four nearest first color components and the two next-nearest first color components on the left.
  • FIG16 shows a schematic diagram of a sampling window for downsampling the first color component in one embodiment of the present application.
  • the blocks arranged in an array represent first color components, including 16011-16016, and the five-pointed stars distributed within the array represent second color components, including 16021 and 16022.
  • a sampling window 1603 having a specified window size corresponding to the pixel position of the second color component 16021 can be determined based on the pixel position of the second color component 16021.
  • the sampling window 1603 can cover the four first color components 16011-16014 that are the closest neighbors to the second color component 16021, as well as the two first color components 16015 and 16016 that are the next closest neighbors to the left.
  • the reconstructed value of the first color component of the target area is downsampled in a sampling window to obtain a reconstructed value of the first color component that matches the pixel position, including: obtaining the correspondence between the reconstructed value of the first color component of the target area and the pixel position in the sampling window; and performing a weighted operation on the reconstructed value of the first color component according to the correspondence to obtain a reconstructed value of the first color component that matches the pixel position.
  • L and L′ represent the reconstructed value of the original first color component and the reconstructed value of the downsampled first color component, respectively.
  • x and y represent the horizontal position coordinate and the vertical position coordinate of the first color component, respectively.
  • the two first color components 16011 and 16013 located on the left side of the second color component 16021 have coordinates of (x, y) and (x, y+1), and can be assigned a weight coefficient of 1/4;
  • the other four first color components 16012, 16014-16016 have coordinates of (x-1, y), x-1, y+1), (x+1, y) and (x+1, y+1), and can be assigned a weight coefficient of 1/8.
  • the two first color components 16015 and 16016 of the next nearest neighbors on the left are unavailable, and the two first color components with coordinates (x-1, y) and (x-1, y+1) are unavailable
  • the two first color components 16011 and 16013 of the nearest neighbors on the left can be used instead, that is, the two first color components with coordinates (x, y) or (x, y+1) can be used instead.
  • the first color component of at least one image area in the reference area when downsampling the first color component of the target area during prediction model training, can be simultaneously downsampled. Specifically, within at least one image area in the reference area, a sampling window having a specified window size is determined based on the pixel locations of the second color component of a second image block, where the reference area includes at least one image area adjacent to the second image block. Within the sampling window, the reconstructed values of the first color component of the at least one image area are downsampled to obtain reconstructed values of the first color component that match the pixel locations.
  • the first color component may be downsampled, or downsampling may not be performed, and the first color component having the same or corresponding position as the second color component may be selected for color component prediction.
  • the first color component 1601 having a position coordinate (x, y) corresponding to the second color component 1602 may be selected for color component prediction.
  • the first color component when the sampling format of the first color component and the second color component is the same, before the color component is predicted by the prediction model, the first color component can be filtered to reduce redundant information or interference information in the first color component, thereby further improving the prediction accuracy of the second color component, that is, improving the video encoding and decoding accuracy of the second image block.
  • multiple filters can be selected at the same time to filter the sample data of the first color component, and then different first color component samples can be selected to train the fitting prediction model to improve the fitting training effect of the prediction model.
  • the pixel position of the first color component can be boundary-extended as needed to obtain neighboring pixel points for the designated pixel point.
  • the reference area includes a variety of different candidate area templates
  • the prediction model also includes a variety of different candidate prediction models.
  • different color component prediction schemes may be selected for image blocks with different block sizes.
  • a plurality of different color component prediction schemes may be used for a current block, and then the prediction results of the plurality of schemes may be weighted to obtain a final color component prediction result.
  • the final predicted image can be obtained by weighting the cross-component prediction results provided by the above embodiments and the prediction results of the common intra prediction mode, inter prediction mode or intra block copy prediction mode.
  • a preset prediction mode for example, intra-frame prediction, inter-frame prediction or intra-block copy prediction is performed on the second image block to obtain an initial prediction value of the second color component of the second image block; after mapping the reconstructed value of the first color component of the target area according to the mapping relationship to obtain the prediction value of the second color component of the second image block, a weighted operation is performed on the initial prediction value and the prediction value of the second color component according to a preset weight coefficient to obtain an updated prediction value of the second color component.
  • the luminance component reconstruction value of the luminance block and the luminance component reconstruction value of the reference area may be downsampled, for example, using the sampling window shown in FIG16 .
  • pixels whose horizontal and vertical coordinates are not odd numbers are selected from the reference area as sample pixels to construct sample pairs including the reconstructed luma component value and the reconstructed chroma component value.
  • pixels whose horizontal and vertical coordinates are at least one even number may be selected from the reference area as sample pixels.
  • the values are input into the first prediction model and the second prediction model that have been fitted and trained, respectively, to obtain prediction results of the chrominance component U and the chrominance component V.
  • the prediction results of the chrominance component U and the chrominance component V are weighted with the initial prediction results according to the following formula to obtain the final prediction results pred of the chrominance component U and the chrominance component V.
  • pred 1 represents the predicted value of the chrominance component U or the chrominance component V obtained by predicting the luminance component Y
  • pred 2 represents the initial predicted value of the chrominance component U or the chrominance component V obtained by predicting the inter-frame prediction mode or the intra-frame block copy prediction mode
  • w represents the weight coefficient for weighted operation of the two prediction results, which can be, for example, 0.75.
  • the cross-component prediction scheme provided by the embodiment of the present application is allowed to be used in either the luma-chroma merged tree or the luma-chroma separate tree.
  • cross-component prediction is allowed if the number of non-zero coefficients of the chroma corresponding luma block is greater than or equal to 3; otherwise, cross-component prediction is prohibited.
  • the number of non-zero coefficients of the corresponding luma block is not limited.
  • the embodiment of the present application proposes a cross-component color component prediction method, which generates a cross-component prediction model based on the predicted image of the adjacent reconstructed pixels or the current block by utilizing the similarity of the mapping relationship between the components of the adjacent areas or the predicted image and the reconstructed image, and then uses the cross-component prediction model to generate the target predicted image of the current image block.
  • the embodiment of the present application can further utilize the reconstructed pixels of the current block to improve the prediction accuracy of the color component, thereby improving the encoding and decoding efficiency.
  • the embodiment of the present application expands the scope of use of cross-component prediction, that is, the luminance and chrominance separation tree mode can also adopt a cross-component color component prediction scheme.
  • cross-component prediction is not allowed, so as to achieve the effect of saving bit rate and improving encoding and decoding efficiency.
  • the following describes an apparatus embodiment of the present application, which can be used to execute the video encoding and decoding method in the above-mentioned embodiment of the present application.
  • the segmentation module 1710 is configured to segment the current video frame to obtain at least one first image block and a second image block, wherein the first image block is a coding unit having a first color component and the second image block is a coding unit having a second color component;
  • the prediction module 1730 is configured to predict and obtain a reconstructed value of the first color component of the target area according to a preset prediction mode
  • the mapping module 1740 is configured to map the reconstructed value of the first color component of the target area according to the mapping relationship between the first color component and the second color component in response to a preset prediction condition being met, so as to obtain a predicted value of the second color component of the second image block.
  • the segmentation module 1710 is configured to segment the current video frame based on a single segmentation tree structure to obtain the at least one first image block and the second image block;
  • the prediction condition includes: the number of non-zero coefficients of the first color component of the target area is greater than a preset threshold value.
  • the prediction conditions include one or more of the following conditions:
  • the first segmentation tree structure is different from the second segmentation tree structure
  • the number of non-zero coefficients of the first color component of the target area is greater than a preset number threshold
  • the sum of the non-zero coefficients of the first color components of the plurality of first image blocks corresponding to the target area is greater than a preset threshold.
  • the prediction conditions include one or more of the following conditions:
  • an index corresponding to the second image block having a specified index value, where the index is used to indicate whether to perform: mapping the reconstructed value of the first color component to obtain a predicted value of the second color component of the second image block;
  • the image feature of the target area satisfies a preset first feature condition
  • the image feature of the second image block satisfies a preset second feature condition
  • the image feature of the reference area corresponding to the second image block meets a preset third feature condition, and the reference area includes at least one image area adjacent to the second image block.
  • the first characteristic condition includes: a reconstruction parameter of the first color component of the target area is within a preset value range.
  • the second characteristic condition includes one or more of the following conditions:
  • the image size of the second image block is within a preset size range
  • the position of the second image block in the current video frame is within a specified area.
  • the area of the reference region corresponding to the second image block is greater than a preset area threshold
  • the number of pixels at the designated sampling position in the reference area corresponding to the second image block is greater than a preset number threshold.
  • the mapping relationship is represented by a prediction model
  • the quantity threshold is the number of model parameters in the prediction model
  • the prediction device 1700 further includes:
  • a weighting module 1750 configured to predict an initial prediction value of the second color component of the second image block according to a preset prediction mode, where the prediction mode includes an intra prediction mode, an inter prediction mode, or an intra block copy prediction mode;
  • a weighted operation is performed on the initial prediction value and the prediction value of the second color component to obtain an updated prediction value of the second color component.
  • mapping relationship is represented by a prediction model
  • prediction device 1700 further includes:
  • an initial model acquisition module 1760 configured to acquire an initial prediction model, wherein an input item of the initial prediction model includes the first color component of a specified pixel point, an output item of the initial prediction model is the second color component of the specified pixel point, and the initial prediction model includes at least one of a plurality of candidate models;
  • the sample pair selection module 1770 is configured to obtain training sample pairs, wherein the training sample pairs include one or more of the following sample pairs:
  • the model fitting module 1780 is configured to perform parameter fitting on the initial prediction model according to the training sample pairs to obtain the prediction model.
  • the predicted value of the first color component of the sample pixel point located in the target area and the predicted value of the second color component of the sample pixel point located in the second image block are predicted according to a preset prediction mode, and the prediction mode includes an intra-frame prediction mode, an inter-frame prediction mode, or an intra-block copy prediction mode.
  • the input item of the initial prediction model also includes the first color component of at least one neighborhood pixel point, and the neighborhood pixel point is the pixel point that is the nearest neighbor or the second nearest neighbor of the designated pixel point.
  • the initial prediction model includes at least one combination item with an independent weighting parameter, and the combination item takes the first color components of at least two pixels as input items.
  • the initial prediction model includes at least two combination terms with different orders, and the order is the highest power of the input terms in the combination terms.
  • the reference area is composed of at least one nearest neighbor area or second nearest neighbor area
  • the nearest neighbor area includes an image area with a specified image size located above or to the left of the second image block
  • the second nearest neighbor area includes an image area with a specified image size located above, below, or above the second image block.
  • next-nearest neighboring region at the lower left of the second image block has the same image size as the second image block in the vertical direction, and the next-nearest neighboring region at the upper right of the second image block has a specified image size in the horizontal direction;
  • the next neighboring area at the upper left of the second image block has a specified image size in both the horizontal direction and the vertical direction.
  • the reference area includes at least one of a full area combination, a left area combination, and an upper area combination;
  • the full region combination includes the nearest neighbor regions located on the left and above the second image block and the next nearest neighbor regions located on the upper left, lower left, and upper right of the second image block;
  • the left region combination includes a nearest neighbor region located on the left side of the second image block and a next nearest neighbor region located on the lower left side of the second image block;
  • the upper region combination includes a nearest neighbor region located above the second image block and a next nearest neighbor region located to the upper right of the second image block.
  • the prediction device 1700 further includes:
  • the area adjustment module 1790 is configured to obtain availability information of at least one sub-area constituting the reference area; and adjust the area range of the reference area according to the availability information.
  • the area adjustment module 1790 is configured to remove sub-areas in an unavailable state from the reference area; when all sub-areas in the reference area are in an unavailable state, the reference area is configured to be in an unavailable state.
  • the sample pair acquisition module 1770 is configured to: acquire a pixel sampling pattern of the second image block, where the pixel sampling pattern includes full pixel sampling or partial pixel sampling;
  • the sample pixel points located in the second image block refer to all pixel points in the second image block
  • the sample pixel points located in the reference area refer to all pixel points in at least one image area in the reference area
  • the sample pixel points located in the second image block refer to the partial pixel points at the specified sampling position
  • the sample pixel points located in the reference area refer to the partial pixel points at the specified sampling position in at least one image area in the reference area.
  • the designated sampling location satisfies at least one of the following conditions:
  • the coordinates of the designated sampling position meet preset coordinate value conditions
  • the reconstructed value of the first color component of the pixel point at the designated sampling position is within a preset value range.
  • the coordinate numerical conditions include:
  • At least one of the horizontal position coordinate and the vertical position coordinate is an even number; or,
  • At least one of the horizontal position coordinate and the vertical position coordinate is an odd number.
  • the prediction device 1700 further includes:
  • a first downsampling module 1791 is configured to determine a sampling window with a specified window size based on the pixel positions of the second color component of the second image block; downsample the reconstructed value of the first color component of the target area within the sampling window to obtain a reconstructed value of the first color component that matches the pixel positions;
  • the mapping module 1740 is configured to map the matched reconstructed value of the first color component to obtain a predicted value of the second color component of the second image block.
  • the first downsampling module 1791 is configured to: obtain, in the sampling window, a correspondence between a reconstructed value of the first color component of the target area and the pixel position;
  • a weighted operation is performed on the reconstructed value of the first color component of the target area to obtain a reconstructed value of the first color component that matches the pixel position.
  • the prediction device 1700 further includes:
  • the second downsampling module 1792 is configured to determine a sampling window with a specified window size in at least one image area in the reference area corresponding to the second image block according to the pixel position of the second color component of the second image block; in the sampling window, downsample the reconstructed value of the first color component of the at least one image area to obtain a reconstructed value of the first color component that matches the pixel position.
  • FIG18 schematically shows a structural block diagram of an electronic device for implementing an embodiment of the present application.
  • electronic device 1800 includes a central processing unit (CPU) 1801, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1802 or programs loaded from a storage unit 1808 into a random access memory (RAM) 1803. Various programs and data required for system operation are also stored in the random access memory 1803.
  • the CPU 1801, the read-only memory 1802, and the random access memory 1803 are connected to each other via a bus 1804.
  • An input/output interface 1805 i.e., an I/O interface
  • I/O interface is also connected to the bus 1804.
  • the following components are connected to the input/output interface 1805: an input section 1806 including a keyboard, a mouse, and the like; an output section 1807 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 1808 including a hard disk and the like; and a communication section 1809 including a network interface card such as a local area network card or a modem.
  • the communication section 1809 performs communication processing via a network such as the Internet.
  • a drive 1810 is also connected to the input/output interface 1805 as needed.
  • a removable medium 1811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like, is installed in the drive 1810 as needed so that a computer program read therefrom can be installed into the storage section 1808 as needed.
  • an embodiment of the present application includes a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for executing the methods shown in the flow charts.
  • the computer program can be downloaded and installed from a network via the communication section 1809 and/or installed from a removable medium 1811.
  • the central processing unit 1801 the various functions defined in the system of the present application are performed.
  • the computer-readable storage medium shown in the embodiments of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above.
  • the computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above.
  • Computer-readable storage media may include, but are not limited to: an electrical connection with at least one wire, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
  • a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device or device.
  • a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.
  • a computer-readable signal medium may also be any computer-readable storage medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.
  • Program code embodied on a computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.
  • each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code includes at least one executable instruction for realizing the specified logical function.
  • the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved.
  • each box in the block diagram or flowchart, and the combination of the boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
  • the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.
  • a non-volatile storage medium which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.
  • a computing device which can be a personal computer, a server, a touch terminal, or a network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

本申请实施例公开了一种图像处理方法、装置、存储介质、电子设备以及程序产品。该方法包括：对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，所述第一图像块是具有第一颜色分量的编码单元，所述第二图像块是具有第二颜色分量的编码单元；根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域；根据预设的预测模式，预测得到所述目标区域的所述第一颜色分量的重建值；及，响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。

Description

图像处理方法、装置、存储介质、电子设备及程序产品
本申请要求于2024年3月14日提交中国专利局、申请号为202410297228.5、申请名称为“颜色分量的预测方法、装置、介质及电子设备”的中国专利申请的优先权。
技术领域
本申请属于音视频技术领域，具体涉及一种图像处理方法、图像处理装置、计算机可读存储介质、电子设备以及计算机程序产品。
发明背景
为了适应音视频数据的大规模数据传输，需要在数据发送端，对原始的音视频数据进行编码处理，形成压缩后的数据码流。在将数据码流传输至数据接收端后，再对数据码流进行解码，还原得到预测重建的音视频数据。
然而，传统的音视频编解码方案存在编解码效率低、准确性差等问题。
发明内容
本申请提供一种图像处理方法、图像处理装置、计算机可读存储介质、电子设备以及计算机程序产品，目的在于提高视频编解码的效率和准确性。
一方面，本申请实施例提供了一种图像处理方法，由电子设备执行，该方法包括：
对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，所述第一图像块是具有第一颜色分量的编码单元，所述第二图像块是具有第二颜色分量的编码单元；
根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域；
根据预设的预测模式，预测得到所述目标区域的所述第一颜色分量的重建值；及，
响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
另一方面，本申请实施例提供了一种图像处理装置，该装置包括：
分割模块，被配置为对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，所述第一图像块是具有第一颜色分量的编码单元，所述第二图像块是具有第二颜色分量的编码单元；
确定模块，被配置为根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域；
预测模块，被配置为根据预设的预测模式，预测得到所述目标区域的所述第一颜色分量的重建值；及，
映射模块，被配置为响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
另一方面，本申请实施例提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现如以上技术方案中的图像处理方法。
另一方面，本申请实施例提供了一种电子设备，该电子设备包括：处理器；以及存储器，用于存储所述处理器的可执行指令；其中，所述处理器被配置为执行所述可执行指令来实现如以上技术方案中的图像处理方法。
另一方面，本申请实施例提供了一种计算机程序产品，包括计算机程序，所述计算机程序存储在计算机可读存储介质中；当电子设备的处理器从所述计算机可读存储介质读取所述计算机程序时，所述处理器执行所述计算机程序，使得所述电子设备执行如以上技术方案中的图像处理方法。
附图简要说明
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本申请的实施例，并与说明书一起用于解释本申请的原理。显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1示意性地示出了可以应用本申请实施例的技术方案的示例性系统架构的示意图。
图2示意性地示出视频编码装置和视频解码装置在流式传输环境中的放置方式。
图3示意性地示出了一个视频编码器的基本流程图。
图4示出了本申请一个实施例中图像处理方法的流程图。
图5示出了本申请一个实施例中对当前块进行颜色分量采样的过程示意图。
图6示意性地示出了多种块划分模式的划分结果。
图7示出了在一个应用场景中对亮度分量和色度分量使用不同的分割树结构进行图像分割得到的图像块。
图8示出了本申请一个实施例中拟合用于表示颜色分量之间的映射关系的预测模型的流程图。
图9示出了本申请一个实施例中指定像素点与邻域像素点的关系示意图。
图10示出了本申请一个实施例中与当前块相对应的参考区域的分布示意图。
图11示出了本申请一个实施例中将部分子区域组成参考区域的区域模板示意图。
图12示出了本申请一个实施例中为当前块选定的参考区域的区域模板示意图。
图13示出了本申请一个实施例中基于像素点的位置坐标选取采样位置的示意图。
图14示出了本申请一个实施例中基于往返扫描方式选取采样位置的示意图。
图15示出了本申请一个实施例中基于ZigZag扫描方式选取采样位置的示意图。
图16示出了本申请一个实施例中对第一颜色分量进行下采样的采样窗口示意图。
图17示意性地示出了本申请实施例提供的图像处理装置的结构框图。
图18示意性示出了适于用来实现本申请实施例的电子设备的结构框图。
实施方式
现在将参考附图更全面地描述示例实施方式。然而，示例实施方式能够以多种形式实施，且不应被理解为限于在此阐述的范例；相反，提供这些实施方式使得本申请将更加全面和完整，并将示例实施方式的构思全面地传达给本领域的技术人员。
此外，所描述的特征、结构或特性可以以任何合适的方式结合在一个或更多实施例中。在下面的描述中，提供许多具体细节从而给出对本申请的实施例的充分理解。然而，本领域技术人员将意识到，可以实践本申请的技术方案而没有特定细节中的一个或更多，或者可以采用其它的方法、组元、装置、步骤等。在其它情况下，不详细示出或描述公知方法、装置、实现或者操作以避免模糊本申请的各方面。
本申请实施例中，术语“模块”或“单元”是指有预定功能的计算机程序或计算机程序的一部分，并与其他相关部分一起工作以实现预定目标，并且可以通过使用软件、硬件(如处理电路或存储器)或其组合来全部或部分实现。同样的，一个处理器(或多个处理器或存储器)可以用来实现至少一个模块或单元。此外，每个模块或单元都可以是包含该模块或单元功能的整体模块或单元的一部分。
附图中所示的方框图仅仅是功能实体，不一定必须与物理上独立的实体相对应。即，可以采用软件形式来实现这些功能实体，或在至少一个硬件模块或集成电路中实现这些功能实体，或在不同网络和/或处理器装置和/或微控制器装置中实现这些功能实体。
附图中所示的流程图仅是示例性说明，不是必须包括所有的内容和操作/步骤，也不是必须按所描述的顺序执行。例如，有的操作/步骤还可以分解，而有的操作/步骤可以合并或部分合并，因此实际执行的顺序有可能根据实际情况改变。
本申请实施例中涉及的相关技术术语解释如下。
视频编码是指处理图片序列，以形成视频或视频序列。在视频编码领域，术语“图片(picture)”、“帧(frame)”或“图像(image)”可以用作同义词。
本申请实施例中使用的视频编码(coding)表示视频编码(encoding)或视频解码(decoding)。其中，视频编码(encoding)在源侧执行，包括处理(例如，压缩)原始视频图像以减少表示该视频图像所需的数据量，从而更高效地存储和/或传输。视频解码在目的地侧执行，相对于编码器作逆处理，以重建视频图像。本申请实施例涉及的对视频帧“编码(coding)”也可以理解为对视频图像序列进行“编码”或“解码”。编码部分和解码部分的组合也称为编解码。
本申请实施例中，视频图像序列中的每个图像被分割成不重叠的块集合，在块层级上进行编码。编码器侧对视频进行编码，也就是指在块(也称为图像块或视频块)层级，包括通过空间(图像内)预测和时间(图像间)预测来产生预测块，从当前块(当前处理的块或待处理的块)减去预测块以获取残差块，在变换域对残差块进行变换并量化，以减少待传输(或待压缩)的数据量。解码器侧将相对于编码器的逆处理部分应用于已编码的或已压缩的块，以重建当前块。
另外，编码器复制解码器侧的循环处理，使得编码器和解码器生成相同的预测(例如帧内预测和帧间预测)块和/或重建块，用于后续块的编码。
术语“块”为图像或帧的一部分。本申请实施例中，当前块是指当前正在处理的块。例如在编码中，指当前正在编码的块；在解码中，指当前正在解码的块。
图1示意性地示出了可以应用本申请实施例的技术方案的示例性系统架构的示意图。
如图1所示，系统架构100包括多个终端装置，所述终端装置可通过例如网络150彼此通信。举例来说，系统架构100可以包括通过网络150互连的第一终端装置110和第二终端装置120。在图1的实施例中，第一终端装置110和第二终端装置120执行单向数据传输。
举例来说，第一终端装置110可对视频数据(例如由终端装置110采集的视频图片流)进行编码，以通过网络150传输到第二终端装置120。已编码的视频数据以至少一个已编码视频码流形式传输。第二终端装置120可从网络150接收已编码视频数据，对已编码视频数据进行解码以恢复视频数据，并根据恢复的视频数据显示视频图片。
在本申请的一个实施例中，系统架构100可以包括第三终端装置130和第四终端装置140，二者可以执行已编码视频数据的双向传输，例如，所述双向传输可以发生在视频会议期间。对于双向数据传输，第三终端装置130和第四终端装置140中的每个终端装置可对视频数据(例如由终端装置采集的视频图片流)进行编码，以通过网络150传输到第三终端装置130和第四终端装置140中的另一终端装置。第三终端装置130和第四终端装置140中的每个终端装置还可接收由第三终端装置130和第四终端装置140中的另一终端装置传输的已编码视频数据，且可对已编码视频数据进行解码以恢复视频数据，并可根据恢复的视频数据在可访问的显示装置上显示视频图片。
在图1的实施例中，第一终端装置110、第二终端装置120、第三终端装置130和第四终端装置140可为服务器、个人计算机和智能电话，但本申请可不限于此。本申请公开的实施例适用于膝上型计算机、平板电脑、媒体播放器和/或专用视频会议设备。网络150表示在第一终端装置110、第二终端装置120、第三终端装置130和第四终端装置140之间传送已编码视频数据的任何数目的网络，包括例如有线和/或无线通信网络。通信网络150可在电路交换和/或分组交换信道中交换数据。该网络可包括电信网络、局域网、广域网和/或互联网。
在本申请的一个实施例中，图2示意性地示出视频编码装置和视频解码装置在流式传输环境中的放置方式。本申请所公开主题可同等地适用于其它支持视频的应用，包括例如视频会议、数字TV(television，电视机)、在包括CD、DVD、存储棒等的数字介质上存储压缩视频等等。
流式传输系统可包括采集子系统213，采集子系统213可包括数码相机等视频源201、视频源创建未压缩的视频图片流202。在实施例中，视频图片流202包括由数码相机拍摄的样本。相较于已编码的视频数据204(或已编码的视频码流204)，视频图片流202被描绘为粗线以强调高数据量的视频图片流，视频图片流202可由电子装置220处理。电子装置220包括耦接到视频源201的视频编码装置203。视频编码装置203可包括硬件、软件或软硬件组合以实现或实施如下文更详细地描述的各个实施例。
相较于视频图片流202，已编码的视频数据204(或已编码的视频码流204)被描绘为细线以强调较低数据量的已编码的视频数据204(或已编码的视频码流204)，其可存储在流式传输服务器205上以供将来使用。至少一个流式传输客户端子系统，例如图2中的客户端子系统206和客户端子系统208，可访问流式传输服务器205以检索已编码的视频数据204的副本207和副本209。
客户端子系统206可包括例如电子装置230中的视频解码装置210。视频解码装置210对已编码的视频数据的传入副本207进行解码，输出视频图片流211，可在显示器212(例如显示屏)或另一呈现装置上呈现。在一些流式传输系统中，可根据某些视频编码/压缩标准对已编码的视频数据204、视频数据207和视频数据209(例如视频码流)进行编码。
应注意，电子装置220和电子装置230可包括图中未示出的其它组件。举例来说，电子装置220可包括视频解码装置，且电子装置230还可包括视频编码装置。
在本申请的一个实施例中，以国际视频编码标准HEVC(High Efficiency Video Coding，H.265)、VVC(Versatile Video Coding，H.266)、以及中国国家视频编码标准AVS(Audio Video Coding Standard)为例，当输入一个视频帧图像之后，会根据一个的块大小，将视频帧图像划分成若干个不重叠的处理单元，每个处理单元将进行类似的压缩操作。这个处理单元被称作CTU(Coding Tree Unit，编码树单元)，或者称之为LCU(Largest Coding Unit，最大编码单元)。继续往下，可以对CTU进行更加精细的划分，得到至少一个基本的编码单元CU，CU是一个编码环节中最基本的元素。
图3示意性地示出了一个视频编码器的基本流程图，在该流程中以帧内预测为例进行说明。
其中，原始图像信号sk[x,y]与预测的图像信号做差值运算，得到残差信号uk[x,y]，残差信号uk[x,y]经过变换及量化处理之后得到量化系数。一方面，通过对该量化系数进行熵编码，得到编码后的比特流；另一方面，通过对该量化系数进行反量化及反变换处理，得到重建的残差信号u′k[x,y]。然后，将预测图像信号与重建的残差信号u'k[x,y]叠加，生成图像信号
一方面，将图像信号输入至帧内模式决策模块和帧内预测模块进行帧内预测处理；另一方面，对图像信号进行环路滤波，输出重建图像信号s'k[x,y]，重建图像信号s′k[x,y]可以作为下一帧的参考图像进行运动估计及运动补偿预测。然后，基于运动补偿预测的结果s′r[x+mx,y+my]和帧内预测结果得到下一帧的预测图像信号并继续重复上述过程，直至编码完成。
在以上视频编码过程中涉及的针对每一个CU的编码操作详细介绍如下。
预测编码(Predictive Coding)：预测编码包括了帧内预测和帧间预测等方式。原始视频信号经过选定的已重建视频信号的预测后，得到残差信号。编码端需要为当前CU决定选择哪一种预测模式(也即编码模式)，并告知解码端。其中，帧内预测(Intra Prediction)是指预测的信号来自于同一图像内已经编码重建过的区域；帧间预测(Inter Prediction)是指预测的信号来自已经编码过的、不同于当前图像的其它图像(称之为参考图像)。
变换及量化(Transform&Quantization)：残差信号经过DFT(Discrete Fourier Transform，离散傅里叶变换)、DCT(Discrete Cosine Transform，离散余弦变换)等变换操作后，将信号转换到变换域中，称之为变换系数。变换系数进一步进行有损的量化操作，会丢失掉一定的信息，使得量化后的信号有利于压缩表达。在一些视频编码标准中，可能有多于一种变换方式可以选择，因此编码端也需要为当前CU选择其中的一种变换方式，并告知解码端。量化的精细程度由量化参数(Quantization Parameter，简称QP)来决定，QP取值较大，表示更大取值范围的系数将被量化为同一个输出，因此会带来更大的失真及较低的码率；相反，QP取值较小，表示较小取值范围的系数将被量化为同一个输出，因此会带来较小的失真，同时对应较高的码率。
熵编码(Entropy Coding)或统计编码：量化后的变换域信号将根据各个值出现的频率进行统计压缩编码，最后输出二值化(0或者1)的压缩码流。同时，编码产生的其他信息，例如选择的预测模式、运动矢量数据等，也需要进行熵编码以降低码率。统计编码是一种无损的编码方式，可以有效的降低表达同样信号所需要的码率，常见的统计编码方式有变长编码(Variable Length Coding，简称VLC)或者基于上下文的二值化算术编码(Content Adaptive Binary Arithmetic Coding，简称CABAC)。
基于上下文的自适应二进制算术编码(CABAC)过程主要包含3个步骤：二进制化、上下文建模和二进制算术编码。在对输入的语法元素进行二值化后，可以通过常规编码模式和旁路编码模式(bypass)对二元数据进行编码。
旁路编码模式(Bypass Coding Mode)，它无须为每个二元位分配特定的概率模型，输入的二元位(bin)值直接用一个简单的旁路编码器进行编码，以加快整个编码以及解码的速度。一般情况下，不同的语法元素之间并不是完全独立的，且相同语法元素自身也具有一定的记忆性。因此，根据条件熵理论，利用其他已编码的语法元素进行条件编码，相对于独立编码或者无记忆编码能够进一步提高编码性能。这些用来作为条件的已编码符号信息称为上下文。
在常规编码模式中，语法元素的二元位顺序地进入上下文模型器。编码器根据先前编码过的语法元素或二元位的值，为每一个输入的二元位分配合适的概率模型，该过程即为上下文建模。通过ctxIdxInc(context index increment，上下文索引增量)和ctxIdxStart(context index Start，上下文起始索引)即可定位到语法元素所对应的上下文模型。将bin值和分配的概率模型一起送入二元算术编码器进行编码后，需要根据bin值更新上下文模型，也就是编码中的自适应过程。
环路滤波(Loop Filtering)：经过变化及量化的信号会通过反量化、反变换及预测补偿的操作获得重建图像。重建图像与原始图像相比，由于存在量化的影响，部分信息与原始图像有所不同，即重建图像会产生失真(Distortion)。因此，可以对重建图像进行滤波操作，例如去块效应滤波(Deblocking filter，简称DB)、SAO(Sample Adaptive Offset，自适应像素补偿)或者ALF(Adaptive Loop Filter，自适应环路滤波)等滤波器，可以有效降低量化所产生的失真程度。由于这些经过滤波后的重建图像，将作为后续编码图像的参考，来对将来的图像信号进行预测，因此上述的滤波操作也被称为环路滤波，即在编码环路内的滤波操作。
基于上述的编码过程，在解码端针对每一个CU，在获取到压缩码流(即比特流)之后，进行熵解码获得各种模式信息及量化系数。然后，对量化系数进行反量化及反变换处理，得到残差信号。另一方面，根据已知的预测模式信息，可获得该CU对应的预测信号，然后将残差信号与预测信号相加之后即可得到重建信号，重建信号再经过环路滤波等操作，产生最终的输出信号。
下面结合具体实施方式对本申请提供的图像处理方法、图像处理装置、计算机可读存储介质、电子设备以及计算机程序产品等技术方案做出详细说明。
图4示出了本申请一个实施例中图像处理方法的流程图，该图像处理方法可以由电子设备执行，例如，由发送或者接收视频编码数据的终端设备或者服务器执行，本申请实施例以终端设备执行的方法作为示例进行说明，该终端设备例如可以是图2所示的视频解码装置210或者视频编码装置220。
如图4所示，本申请实施例中的图像处理方法包括如下的步骤S410至S440。
S410：对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，第一图像块是具有第一颜色分量的编码单元，第二图像块是具有第二颜色分量的编码单元；
S420：根据第二图像块在当前视频帧中的位置，将至少一个第一图像块中与位置关联的编码区域，确定为目标区域；
S430：根据预设的预测模式，预测得到目标区域的第一颜色分量的重建值；及，
S440：响应于预设的预测条件被满足，根据第一颜色分量与第二颜色分量之间的映射关系，对第一颜色分量的重建值进行映射，得到第二图像块的第二颜色分量的预测值。
在本申请实施例中，S430中预设的预测模式包括帧内预测模式、帧间预测模式或者帧内块复制预测模式，在使用这些预测模式时用到的预测像素，是通过在已重建的参考帧或者当前帧的已重建区域进行搜索或者在其亚像素(Subpixel)的插值图像中搜索得到。
相关技术中，对于色度分量，一般默认其与亮度分量具有类似的运动矢量或者块矢量，所以色度的运动矢量或者块矢量是基于亮度的运动矢量或者块矢量简单推导得到的。由于这个匹配搜索的过程不仅要考虑到图像的匹配度，还需要考虑到运动矢量或者块矢量的编码代价，因此预测精度仍然存在一定的上升空间。
本申请实施例采用颜色分量映射预测的方式，对第二颜色分量进行预测，实现了针对色度的跨分量预测，克服了传统编码方案中对不同颜色分量独立预测而导致预测精度差的问题，使得跨分量预测能够在亮度色度分离树中正常使用，扩大了跨分量预测的使用范围，提升了跨分量预测方法的性能，因而能够提高视频的编解码效率和准确性。
在S410中，对于彩色的视频帧，在颜色空间中，图像块可以对应一种颜色分类。
根据本申请的一个实施例，可以在颜色空间中，对当前块进行颜色分量采样，可以形成亮度块和色度块。
彩色视频中的图像除了含有亮度分量(Y)以外，还含有色度分量(U，V)，这样的图像又可称为YUV图像。对YUV图像编码时，除了对亮度分量进行编码，还需要对色度分量进行编码。由于人眼对亮度比较敏感，而对色彩不怎么敏感，在编码中为了节省存储空间，提高编码效率，可以对亮度分量进行全分辨率采样，而对色度分量，可不用全分辨率采样。按照彩色视频中亮度分量和色度分量的采样方法的不同，一般视频序列的图像有4:4:4格式的YUV图像、4:2:2格式的YUV图像、4:2:0格式的YUV图像等等。
图5示出了本申请一个实施例中对当前块进行颜色分量采样的过程示意图。如图5所示，当前块501是一个尺寸为16×16的图像块，在对其进行颜色分量采样时，可以选择不同的分辨率格式进行采样。
4:4:4格式表示没有对色度分量进行下采样，4:4:4格式是色度分量分辨率最高的格式。当采用4:4:4格式进行采样时，一个Y分量对应一组UV分量，在相邻4个像素点里的数据有4个Y、4个U、4个V。
4:2:2格式表示色度分量相对于亮度分量进行2:1的水平下采样，没有竖直下采样。对于每两个U采样点或V采样点，每行都包含四个Y采样点。当采用4:2:2格式进行采样时，每两个Y分量共用一组UV分量，在相邻的4个像素点里的数据有4个Y、2个U、2个V。
4:2:0格式表示色度分量相对于亮度分量进行2:1的水平下采样，以及2:1的竖直下采样。4:2:0格式是色度分量分辨率最低的格式。在4:2:0格式中，色度采样在每行中(即水平方向)只有亮度采样的一半，在每列中(即竖直方向)只有亮度采样的一半。当采用4:2:0格式进行采样时，U分量与Y分量交替出现，例如YUV三个分量在第一行中的出现比例为4:2:0，那么在第二行中的出现比例则为4:0:2；在相邻的4个像素点里有4个Y、1个U和1个V。
在视频图像采用4:2:0格式的情况下，若图像块的亮度分量为2M×2N大小的图像块，则图像块的色度分量为M×N大小的图像块。举例来说，如果图像块的分辨率是720*480，那么图像块的亮度分量的分辨率是720*480，图像块的色度分量的分辨率是360*240。
本申请实施例以4:2:0格式为例，对当前块501进行采样后，可以得到一个亮度块502和色度块503，亮度块502为一个16×16的图像块，与之对应的色度块503为一个8×8的图像块。504所示的是与亮度块502和色度块503近邻的图像区域。
在S410中，对当前视频帧进行分割，可以得到多个图像区域，每个图像区域或者多个图像区域的组合可以视为一个图像块。
在本申请的一个实施例中，第一颜色分量的图像块与第二颜色分量的图像块，可以按照相同的分割树结构(也就是单个分割树结构)进行图像分割，形成合并模式的图像块，或者也可以按照不同的分割树结构进行图像分割，形成分离模式的图像块。其中，第一颜色分量可以是亮度分量Y、色度分量U和色度分量V中的任意一种，第二颜色分量可以是亮度分量Y、色度分量U和色度分量V中的另一种。
举例而言，AVS3中的分割树结构一般由四叉树、二叉树和扩展四叉树组成。当块比较大的时候，亮度块和色度块遵循同样的块划分结构，亮度块和色度块是一一对应的，即一个色度块只对应到一个亮度块。而当块划分到比较小的时候，亮度和色度将独立进行块划分，亮度块和色度块将不再一一对应，即一个色度块对应的亮度区域可能在不同的亮度块中。
视频信号中的一帧图像，可以被分割成互不重叠的编码树单元CTU。例如，一个CTU的块尺寸可以设置为64×64、128×128等。示例性地，块尺寸为128×128的CTU可以是包含128列的矩阵像素点阵，每一列包含128个像素，每个像素包含亮度分量和/或色度分量。继续往下，可以对CTU进行更细粒度的划分，得到一组编码单元CU，一组CU里面可以包含至少一个CU。
对编码树单元CTU的划分，可以采用QT(Quaternary Tree，四叉树)、BT(Binary Tree，二叉树)和TT(Ternary Tree，三叉树)等划分方式中的一种划分方式或多种划分方式的组合。其中，按照不同的划分方向，BT可以进一步包括HBT(Horizontal Binary Tree，横向二叉树)、VBT(Vertical Binary Tree，竖向二叉树)，TT可以进一步包括HTT(Horizontal Ternary Tree，横向三叉树)和VTT(Vertical Ternary Tree，竖向三叉树)。
针对某一编码树单元CTU，可以将该CTU作为根节点(root)，划分成若干节点。一个节点对应于一个图像区域。如果对某一节点不再继续划分，则将该节点称为叶节点(Leaf Node)，且其所对应的图像区域就形成一个CU；如果对该节点继续划分，则可以采用上述一种划分方式或多种划分方式的组合，将该节点划分为多个子区域，每个子区域对应一个子节点，之后需要分别确定这些子节点是否还会继续划分。
示例性地，假设根节点的划分层级为0，则子节点的划分层级可以为父节点的划分层级加1。在视频编码过程中，编码器设置有CU的最小块尺寸，在划分过程中，若某一节点的块尺寸等于该最小块尺寸，则该节点默认不再继续划分。为了便于表述，下文中将“节点对应的图像区域”简称为“节点”。针对某一节点，可以采用QT划分模式，将该节点划分为4个子节点。按照QT划分模式，可以将该节点划分成四个相同块尺寸的子节点，各个子节点的宽相同、高也相同，且宽为划分前节点的宽的一半，高为划分前节点的高的一半。
图6示意性地示出了多种块划分模式的划分结果。如图6所示，对于块尺寸为64×64的节点610，若不再继续划分该节点610，则该节点610直接确定为1个块尺寸为64×64的CU；若继续划分该节点610，则可以按照QT划分模式，将该节点610划分为4个块尺寸为32×32的节点621-624。对于这4个块尺寸为32×32的节点621-624中的某一节点，若继续划分该节点，则产生4个块尺寸为16×16的节点。
针对某一节点，可以采用BT划分模式，将该节点划分成2个子节点。可选地，BT划分模式包括两种类型：HBT划分模式和VBT划分模式。其中，HBT划分模式是将节点划分成上、下两个相同块尺寸的子节点，各个子节点的宽相同、高也相同，且宽等于划分前节点的宽，高为划分前节点的高的一半。VBT是将节点划分成左、右两个相同大小的子节点，各个子节点的宽相同、高也相同，且宽为划分前节点的宽的一半，高等于划分前节点的高。
例如，如图6所示，对于块尺寸为64×64的节点610，若按照HBT划分模式，可以将该节点610划分为2个块尺寸为64×32的节点631、632，或者按照VBT划分模式，将该节点610划分为2个块尺寸为32×64的节点641、642。
针对某一节点，可以采用TT划分模式将该节点划分为3个子节点。可选地，TT划分模式包括两种类型：HTT划分模式和VTT划分模式。其中，HTT划分模式是将节点划分成上、中、下三个子节点，如图6中的651-653所示，其中，上、下这两个子节点651、653的宽等于划分前节点610的宽、高为划分前节点610的高的四分之一，中这一节点652的宽等于划分前节点的宽、高为划分前节点的高的一半。VTT是将节点划分成左、中、右三个子节点，如图6中的661-663所示，其中，左、右这两个子节点661、663的宽为划分前节点610的宽的四分之一、高等于划分前节点610的高，中这一节点662的宽为划分前节点的宽的一半、高等于划分前节点的宽。
在本申请的一个实施例中，当第一图像块与第二图像块具有相同的分割树结构时，可以基于单个分割树结构，对当前视频帧进行分割，得到至少一个第一图像块和第二图像块。此时，可以基于第二图像块在所述当前视频帧中的位置，将至少一个第一图像块中与第二图像块相对应的一个第一图像块的全部区域，确定为目标区域。也就是说，在至少一个第一图像块中，存在一个完整的第一图像块与第二图像块相对应，那么，将该第一图像块的全部区域，确定为目标区域。
在本申请的另一个实施例中，当第一图像块与第二图像块具有不同的分割树结构时，可以基于第一分割树结构，对当前视频帧进行分割，得到至少一个第一图像块；基于第二分割树结构，对当前视频帧进行分割，得到第二图像块。此时，可以基于第二图像块在所述当前视频帧中的位置，将至少一个第一图像块中与第二图像块相对应的部分区域或者全部区域，确定为目标区域。也就是说，在至少一个第一图像块中，不仅只有一个第一图像块与第二图像块相对应，那么，可以将与第二图像块相对应的至少一个第一图像块中的部分区域或者全部区域，确定为目标区域。
例如，在VTM7中，图像分割方案可以支持一个CTU中的亮度块和色度块具有独立的分割树结构。例如，对于P帧和B帧，一个CTU中的亮度块和色度块共用相同的分割树结构，即亮度色度合并模式。然而，对于I帧，一个CTU中的亮度块和色度块可以具有不同的分割树结构，即亮度色度分离模式。当使用不同的图像分割模式时，可以基于一个编码树结构，将亮度块分割成亮度CU(尺寸可能更小)，而基于另一个编码树结构，将色度块分割成色度CU。这意味着，I帧中的CU可以包括亮度分量的CU或两个色度分量的CU，除非视频是单色的；P帧或B帧中的CU可以始终包括所有三个色彩分量的CU，即亮度分量的CU和两个色度分量的CU。
在一些实施例中，可以采用半解耦树(semi-decoupled tree，SDT)的图像分割方案，其也被称为半分离树(semi-separate tree，SST)或用于色度分量的灵活块分割。在SDT中，一个超级块(super block，SB)中的亮度块和色度块，可以具有相同或不同的块分割方式，这取决于亮度块的块尺寸或亮度树深度。例如，当亮度块的块尺寸大于阈值T1或亮度块的编码树分裂深度小于或等于阈值T2时，则与亮度块所关联的色度块可以使用与亮度块相同的编码树结构。否则，当亮度块的块尺寸小于或等于T1或亮度块的亮度分裂深度大于T2时，所关联的色度块可以与亮度块具有不同的块分割方式。因此，该方案被称为色度分量的灵活块分割。T1是正整数，例如128或256。T2是正整数，例如1或2。
图7示出了在一个应用场景中对亮度分量和色度分量使用不同的分割树结构进行图像分割得到的图像块，其中的数字1、2、3标识了执行分割的次序。其中，T2被设置为1，即当亮度块的亮度分裂深度大于1时，与之关联的色度块采用与之不同的分割数据结构。此时，根据703在当前视频帧中的位置，可以确定703对应了亮度块中的701和702，那么，可以将701和702的部分区域或者全部区域，确定为目标区域。
当采用改进的半解耦分割(semi-decoupled partitioning，SDP)方案时，亮度块和色度块可以从超级块的根节点开始共用一个部分树结构(partial tree structure)。何时开始对亮度块和色度块执行单独的树分割方案，取决于亮度块的分割信息或码流中的高级语法。
在S430中，目标区域的第一颜色分量的重建值，可以根据预设的预测模式预测得到，该预设的预测模式可以包括帧间预测、帧内预测或者帧内块复制预测等方式。视频编码中，主要的冗余信息是时间冗余，其次是空间冗余，视频编码通过帧间预测消除时间冗余，通过帧内预测消除空间冗余。
帧内预测是利用相邻块的重建值进行预测，该过程通过对已编码的相邻像素进行某种预先设置的加权处理，从而得到对当前像素块的较佳估计。帧内预测的具体方法分两种，一维预测和二维预测。一维预测是利用同行相邻像素点的相关性进行预测，基于下一个像素点值和它前一个像素值比较接近，可以将前一个像素值作为当前像素点的预测值。二维预测除了利用本行的相邻像素点进行预测外，还使用前一行的像素点进行预测。通过给不同行的像素值赋予相应的加权值，最后获得预测值。
帧间预测是利用视频序列中前后帧的相关性对当前帧进行预测。在一个视频序列中，相邻帧之间存在较高的像素相似度，因此可以利用前一帧或后一帧的信息来预测当前帧的像素值。帧间预测可以通过运动估计、运动补偿和残差编码等过程来实现。其中，运动估计是通过比较当前帧与前一帧或后一帧的差异，找到最佳的运动矢量来描述两帧之间的运动关系。常用的运动估计算法可以包括全搜索算法、块匹配算法等。运动补偿是利用找到的运动矢量对前一帧或后一帧进行修正，以实现对当前帧的像素预测。运动补偿可以通过将运动矢量应用于参考帧进行像素复制或插值等方式来实现。残差编码是通过将当前帧的像素值与预测帧的像素值之间的差异进行编码，得到残差数据。残差数据表示了当前帧中无法通过运动预测得到的像素信息，需要额外进行编码传输。
帧内块复制预测(Intra Block Copy，IBC)可以视为一种特殊的帧间预测模式，帧内块复制预测模式的实现原理与帧间预测模型中进行运动补偿的方案原理几乎一致。不同之处在于，帧间预测模式中，在与当前视频帧不同的参考视频帧中选择用于进行运动补偿的参考块，而帧内块复制预测模式中，在当前视频帧的内部选择用于进行运动补偿的参考块。在帧内块复制预测模式下，运动矢量表示在当前视频帧内由当前块的位置移动到参考块的位置的相对位移。
在本申请的一个实施例中，可以根据当前块与参考区域中的至少一个图像区域，拟合用于表示第一颜色分量与第二颜色分量的映射关系的预测模型，参考区域包括至少一个与当前块最近邻的或次近邻的已解码的图像区域。
第二颜色分量是在颜色空间中与第一颜色分量不同的另一种颜色分量，例如当第一颜色分量为亮度分量Y时，第二颜色分量可以是色度分量U或者色度分量V。
以YUV颜色空间为例，本申请实施例中，对具有YUV图像格式的当前块，可以在多种颜色分量之间，拟合得到如下所示的多种映射关系：
(1)根据亮度分量Y预测色度分量U，得到映射关系U＝f(Y)。
(2)根据亮度分量Y预测色度分量V，得到映射关系V＝f(Y)。
(3)根据色度分量U预测色度分量V，得到映射关系V＝f(U)
(4)根据亮度分量Y和色度分量U预测色度分量V，得到映射关系V＝f(Y,U)。
(5)根据亮度分量Y和色度分量V预测色度分量U，得到映射关系U＝f(Y,V)。
图8示出了本申请一个实施例中拟合用于表示颜色分量之间的映射关系的预测模型的流程图。如图8所示，在以上实施例的基础上，预测模型的获取方法可以包括如下的步骤S810至S830。
S810：获取初始预测模型，初始预测模型的输入项包括指定像素点的第一颜色分量，初始预测模型的输出项为指定像素点的第二颜色分量，初始预测模型包括多种候选模型中的至少一种。
在本申请的一个实施例中，初始预测模型的输入项还包括至少一个邻域像素点的第一颜色分量，邻域像素点是与指定像素点最近邻或次近邻的像素点。
图9示出了本申请一个实施例中指定像素点与邻域像素点的关系示意图。
如图9所示，C代表指定像素点，第一颜色分量a和第二颜色分量b中的指定像素点可以是关联位置或者相同位置的像素样本点。
指定像素点C的邻域像素点可以包括最近邻的多个像素点，例如图9所示的位于指定像素点C上方、下方、左侧、右侧位置的多个像素点N、S、W和E。
指定像素点C的邻域像素点可以包括次近邻的多个像素点，例如图9所示的位于指定像素点C左上、右上、左下和右下位置的多个像素点NW、NE、SW和SE。
第一颜色分量a和第二颜色分量b每建立这样一个连接，可以称为跨分量匹配对，即可生成一个映射关系等式。多个这样的映射关系等式即可求解模型的加权参数。图9中指定像素点C的位置仅为举例，也可以选择更大范围中的像素样本点。
在本申请的一个实施例中，初始预测模型包括至少一个具有独立的加权参数的组合项，组合项以至少两个像素点的第一颜色分量作为输入项。
本申请实施例中的初始预测模型可以由至少一个单项式组成，其中每个单项式可以具有独立的加权参数。当一个单项式中存在至少两个像素点的第一颜色分量作为输入项时，该单项式即称为组合项。
在本申请的一个实施例中，初始预测模型可以包括至少两个具有不同阶数的组合项，阶数是组合项中输入项的最高次幂。通过次幂运算可以引入非线性因子，提高模型对于颜色分量映射关系的拟合效果。
在本申请的一个实施例中，初始预测模型可以由以下单项式中的至少一个进行加权运算得到：x，mx±ny，xy，xk，(mx±ny)k，(mx±ny)(pz±qf)，(mx±ny)x，B。
其中，m、n、p和q表示用于对输入项进行固定加权的加权系数；x、y、z和f表示输入项，即指定像素点或者邻域像素点的第一颜色分量；B表示常数偏置项；k表示对输入项进行次幂运算的阶数，k为大于1的整数。
在本申请的一个实施例中，结合图9所示，初始预测模型可以包括如下多个候选模型中的至少一个：
(1)Cb＝p0C+p1N+p2S+p3W+p4E+p5C2+p6B。
(2)Cb＝p0C+p1N+p2S+p3W+p4E+p5B。
(3)Cb＝p0C+p1B。
(4)
(5)
(6)
(7)
(8)
(9)
(10)p6B。
(11)
(12)Cb＝p0C+p1S+p2W+p3E+p4SW+p5SE+p6C2+p7B。
其中，p0、p1、p2、p3、p3、p4、p5、p6、p7表示权重。
S820：获取训练样本对，训练样本对包括以下样本对中的一种或多种：
位于目标区域内的样本像素点的第一颜色分量的预测值和位于第二图像块内的样本像素点的第二颜色分量的预测值、
位于参考区域内的样本像素点的第一颜色分量的重建值和第二颜色分量的重建值，参考区域包括至少一个与第二图像块相邻的图像区域。
在本申请的一个实施例中，可以根据第二图像块和对应的目标区域自身的颜色分量的预测值，拟合第一颜色分量与第二颜色分量的映射关系。此时，用于拟合映射关系的训练样本对，可以包括(位于目标区域内的样本像素点的第一颜色分量的预测值，位于第二图像块内的样本像素点的第二颜色分量的预测值)。
在本申请的一个实施例中，位于目标区域内的样本像素点的第一颜色分量的预测值和位于第二图像块内的样本像素点的第二颜色分量的预测值，是根据预设的预测模式预测得到的，该预设的预测模式包括帧内预测模式、帧间预测模式或者帧内块复制预测模式。
在本申请的一个实施例中，可以根据参考区域的颜色分量的重建值，拟合第一颜色分量与第二颜色分量的映射关系。此时，用于拟合映射关系的训练样本对，可以包括：(位于参考区域内的样本像素点的第一颜色分量的重建值，该样本像素点的第二颜色分量的重建值)。
在本申请的一个实施例中，可以同时根据第二图像块和目标区域自身的颜色分量的预测值以及参考区域的颜色分量的重建值，拟合第一颜色分量与第二颜色分量的映射关系。此时，用于拟合映射关系的训练样本对，可以包括上述二种，即：(位于目标区域内的样本像素点的第一颜色分量的预测值，位于第二图像块内的样本像素点的第二颜色分量的预测值)，以及，(位于参考区域内的样本像素点的第一颜色分量的重建值，该样本像素点的第二颜色分量的重建值)。
在本申请的一个实施例中，与当前块(如无特殊说明，以下所述的当前块均表示第二图像块)相对应的参考区域由至少一个最近邻区域或次近邻区域组合而成，最近邻区域包括位于当前块上方或左侧的具有指定图像尺寸的图像区域，次近邻区域包括位于当前块左上方、左下方或右上方的具有指定图像尺寸的图像区域。
图10示出了本申请一个实施例中与当前块相对应的参考区域的分布示意图。
如图10所示，与当前块1001相对应的参考区域可以由多个子区域1002组成，其中每个子区域1002可以是当前块的最近邻区域或次近邻区域。其中，最近邻区域例如可以包括位于当前块1001上方的图像区域B或者位于当前块1001左侧的图像区域D，次近邻区域例如可以包括位于当前块1001左上方的图像区域A、位于当前块1001左下方的图像区域E或者位于当前块1001右上方的图像区域C。
由图像区域A-E中的至少一个进行组合，可以形成当前块1001的参考区域。
在本申请的一个实施例中，用于组成参考区域的各个子区域的图像尺寸规定如下：第二图像块右上方的次近邻区域在水平方向上与第二图像块具有相同的图像尺寸，第二图像块右上方的次近邻区域在竖直方向上具有指定图像尺寸；
第二图像块左下方的次近邻区域在竖直方向上与第二图像块具有相同的图像尺寸，第二图像块右上方的次近邻区域在水平方向上具有指定图像尺寸；
第二图像块左上方的次近邻区域在水平方向上和竖直方向上均具有指定图像尺寸。
具体地，参见图10，当前块1001上方的最近邻区域B在水平方向上与当前块1001具有相同的图像尺寸，当前块1001上方的最近邻区域B在竖直方向上具有指定图像尺寸。
当前块1001左侧的最近邻区域D在竖直方向上与当前块1001具有相同的图像尺寸，当前块1001左侧的最近邻区域D在水平方向上具有指定图像尺寸。
当前块1001右上方的次近邻区域C在水平方向上与当前块1001具有相同的图像尺寸，当前块1001右上方的次近邻区域C在竖直方向上具有指定图像尺寸。
当前块1001左下方的次近邻区域E在竖直方向上与当前块1001具有相同的图像尺寸，当前块1001左下方的次近邻区域在水平方向上具有指定图像尺寸。
当前块1001左上方的次近邻区域A在水平方向上和竖直方向上均具有指定图像尺寸。
指定图像尺寸中的宽和/或高可以是大于或等于一的预设值，例如可以将指定图像尺寸的宽设置为6。当指定图像尺寸宽和/或高大于一时，可以利用多层邻域像素点进行颜色分量预测，因而能够提高颜色分量预测的准确性。
在本申请的一个实施例中，组成当前块的参考区域的各个子区域，可以具有相同的指定图像尺寸或者不同的指定图像尺寸。例如，图像区域C的竖直方向尺寸可以与图像区域E的水平方向尺寸相同或者不同。
在本申请的一个实施例中，可以从子区域中选取已完成解码重建的像素点或者允许可用的像素点，作为当前块进行颜色分量预测的参考像素点。例如，当图像区域C中的一部分像素点已完成解码重建，而另一部分像素点未完成解码重建，此时可以从图像区域C中选取已完成解码重建的像素点作为当前块进行颜色分量预测的参考像素点。
在本申请的一个实施例中，当子区域中已完成解码重建的像素点不满足上述的尺寸规定时，可以将该次近邻区域配置为不可用。例如，当图像区域C的右下角未重建或者超出图像边界时，可以将图像区域C配置为不可用；又例如，当图像区域E的右下角未重建或者超过图像边界时，可以将图像区域E配置为不可用。
在本申请的一个实施例中，可以将图8所示的全部子区域，组合形成当前块的参考区域，或者也可以将其中一部分子区域，组合形成当前块的参考区域。
图11示出了本申请一个实施例中将部分子区域组成参考区域的区域模板示意图。如图11所示，基于不同子区域的组合，可以形成示例的十种候选区域模板，本申请实施例中，可以为当前块指定其中一种或多种候选区域模板。
图12示出了本申请一个实施例中为当前块选定的参考区域的区域模板示意图。如图12所示，在本申请实施例中，与当前块相对应的参考区域包括全量区域组合、左侧区域组合或者上方区域组合中的至少一种。
其中，全量区域组合包括位于当前块左侧和上方的最近邻区域以及位于当前块左上方、左下方和右上方的次近邻区域。
左侧区域组合包括位于当前块左侧的最近邻区域以及位于当前块左下方的次近邻区域。
上方区域组合包括位于当前块上方的最近邻区域以及位于当前块右上方的次近邻区域。
在本申请的一个实施例中，可以在视频码流中使用指示字段来标识当前块使用的参考区域的区域模板。例如，当指示字段取值为1时，表示当前块选用的参考区域为图12中所示的全量区域组合；当指示字段取值为01时，表示当前块选用的参考区域为图12中所示的左侧区域组合；当指示字段取值为00时，表示当前块选用的参考区域为图12中所示的上方区域组合。
在本申请的一个实施例中，在确定当前块进行颜色分量预测时使用的参考区域后，可以获取组成参考区域的至少一个子区域的可用性信息，然后根据至少一个子区域的可用性信息，调整参考区域的区域范围。
在本申请的一个实施例中，根据所述至少一个子区域的可用性信息，调整所述参考区域的区域范围，可以进一步包括：从参考区域中移除处于不可用状态的子区域；当参考区域中的所有子区域均处于不可用状态时，将参考区域配置为不可用状态。
举例而言，通过解析视频码流获得当前块对应的指示字段取值为1，表示当前块进行颜色分量预测时使用的参考区域是图12中所示的包括A-E五个子区域的全量区域组合。
在对当前块进行编码预测时，可以获取参考区域中各个子区域的可用性信息，并根据该可用性信息调整参考区域的区域范围。
例如，子区域A、B、C是位于当前块上方的其他图像块的图像区域，如果子区域A、B、C所在的图像块尚未完成编码重建，则子区域A、B、C处于不可用的状态，此时可以将参考区域的区域范围由A+B+C+D+E调整为D+E。
又例如，子区域D和E是位于当前块左侧的其他图像块的图像区域，如果子区域D和E所在的图像块也尚未完成编码重建，则子区域D和E也处于不可用的状态。此时，A-E五个子区域均处于不可用的状态，因此可以将基于该全量区域组合的参考区域整体配置为不可用状态。
在本申请的一个实施例中，用于训练预测模型的样本像素点可以是在第二图像块与参考区域中的至少一个图像区域内选取的全部像素点或者部分像素点。当选取部分像素点训练预测模型时，可以根据预设的采样规则，在第二图像块与参考区域中的至少一个图像区域内进行像素点采样。
在本申请的一个实施例中，在第二图像块与参考区域中的至少一个图像区域内选取训练样本对，可以包括：
获取第二图像块的像素点采样模式，像素点采样模式包括全量像素点采样或者部分像素点采样；
当所述第二图像块的像素点采样模式为所述全量像素点采样时，所述位于所述第二图像块内的样本像素点是指所述第二图像块内的所有像素点，所述位于所述参考区域内的样本像素点是指所述参考区域中的至少一个图像区域内的所有像素点；
当所述第二图像块的像素点采样模式为所述部分像素点采样时，所述位于所述第二图像块内的样本像素点是指处于指定采样位置的部分像素点，所述位于所述参考区域内的样本像素点是指所述参考区域中的至少一个图像区域内处于指定采样位置的部分像素点。
在本申请的一个实施例中，指定采样位置包括满足以下条件中的至少一种。
第一种采样位置：指定采样位置的坐标，满足预设的坐标数值条件。
在本申请的一个实施例中，坐标数值条件包括：像素点的水平位置坐标或者竖直位置坐标中的至少一个为偶数；或者，像素点的水平位置坐标或者竖直位置坐标中的至少一个为奇数。
图13示出了本申请一个实施例中基于像素点的位置坐标选取采样位置的示意图。
如图13所示，在参考区域中，以左上角作为坐标原点(0，0)建立坐标系，水平位置坐标x表示像素点在水平方向上由左向右排列的顺序位置，竖直位置坐标y表示像素点在水平方向上由上向下排列的顺序位置。
根据预设的坐标数值条件，可以在水平方向上选取偶数位置、奇数位置或者所有位置作为指定采样位置，也可以在竖直方向上选取偶数位置、奇数位置或者所有位置作为指定采样位置。
例如，在图13所示的实施例中，选择水平方向上为偶数位置(即X％2＝＝0)且竖直方向上为所有位置的像素点位置作为指定采样位置，即图中阴影部分的像素点位置选取为指定采样位置。
第二种采样位置：沿预设的像素点扫描方向，选取指定采样位置。
在本申请的一个实施例中，可以通过往返扫描或者ZigZag扫描等任意的扫描方式，在参考区域中扫描像素点，从而沿着预设的像素点扫描方向,按照顺序和预设的采样规则选取指定采样位置。其中，预设的采样规则例如可以是间隔采样，即每间隔至少一个扫描像素点后选取一个指定采样位置。
图14示出了本申请一个实施例中基于往返扫描方式选取采样位置的示意图。
如图14所示，在参考区域中按照从左向右的顺序扫描一行像素点，到达边界后再按照从右向左的顺序扫描下一行像素点。在像素点扫描过程中，沿着扫描方向，每间隔一个扫描像素点选取一个指定采样位置。例如，图中所示的箭头为扫描方向，阴影部分的像素点位置即为选取的指定采样位置。
图15示出了本申请一个实施例中基于ZigZag扫描方式选取采样位置的示意图。
如图15所示，在参考区域中按照ZigZag扫描方式从左上角向右下角扫描像素点。在像素点扫描过程中，沿着扫描方向，每间隔一个扫描像素点选取一个指定采样位置。例如，图中所示的箭头为扫描方向，阴影部分的像素点位置即为选取的指定采样位置。
第三种采样位置：指定采样位置的像素点的第一颜色分量的重建值，处于预设的数值范围内。
以第一颜色分量为亮度分量为例，本申请实施例可以选取亮度值大于或小于某个阈值的像素点所在位置作为指定采样位置。
举例而言，选取参考区域中亮度值大于设定阈值的样本像素点可以拟合得到一个预测模型，选取小于或等于该设定阈值的样本像素点可以拟合得到另一个预测模型；在生成预测图像的时候，可以对亮度值大于设定阈值的像素点使用对应的第一个预测模型，而对亮度值小于或等于设定阈值的像素点使用对应的第二个预测模型。
S830：根据训练样本对，对初始预测模型进行参数拟合，得到所述预测模型。
在本申请的一个实施例中，预测模型可以通过在线训练或者离线训练的方式拟合模型参数。
其中，在线训练指的是在当前视频帧的编解码过程中进行训练拟合，其中每个当前块在进行编解码时可以在线计算模型参数。
离线训练指的是在编解码过程之外离线地拟合计算模型参数。在离线训练过程中，根据预先采集的样本数据集，离线训练好模型参数。然后在编解码时直接使用训练好的目标预测模型，在当前视频帧的编解码过程中不再计算模型参数。与在线训练相比，离线训练的模型预测精度有所降低，但是编解码速度快，因此能够适用于对编解码质量要求不高但对编解码速度要求较高的应用场景中。
在本申请的一个实施例中，在预测模型的训练过程中，可以在当前块与参考区域中的至少一个图像区域内采样得到多个样本像素点，然后根据样本像素点的第一颜色分量和第二颜色分量，对初始预测模型进行参数拟合得到目标预测模型。
举例而言，本申请实施例可以将样本像素点的第一颜色分量和第二颜色分量输入至初始预测模型中，建立多个等式方程，即Ax＝b。
其中，A为M行N列的矩阵，M和N分别表示根据初始预测模型建立的等式数目和初始预测模型中的模型参数数量，x表示以初始预测模型中的模型参数作为元素组成的参数向量。
求解上述等式方程中的参数向量x即可获得所有模型参数pi的值。
在本申请的一个实施例中，可以选取LDL分解法或者高斯消元法求解上述等式方程。以LDL分解法为例，求解步骤可以包括：
(1)将等式方程变换为ATAx＝ATb；
(2)对ATA进行分解，得到LDLTx＝ATb；
(3)求解LY＝ATb得到矩阵Y；
(4)求解DLTx＝Y得到参数向量x，从而获得模型参数pi。
在本申请的一个实施例中，当选取亮度分量Y作为第一颜色分量，选取色度分量U和色度分量V作为第二颜色分量时，即根据亮度分量Y预测色度分量U和色度分量V，由于两个预测过程都需要对ATA进行分解，此时可以合并两个色度分量预测过程中ATA的分解过程，从而降低计算复杂度，提高编解码效率。
基于拟合得到的目标预测模型，可以将目标区域的第一颜色分量的重建值作为输入项，输入至目标预测模型中，由目标预测模型输出第二图像块的第二颜色分量的预测值。
在本申请的一个实施例中，可以对当前视频帧中的一部分图像块执行以上实施例提供的颜色分量预测方案，而对另一部分图像块禁止执行以上实施例提供的颜色分量预测方案。在此基础上，可以显式或者隐式地标识与图像块相对应的预测条件，并在满足预测条件时执行依据第一颜色分量预测第二颜色分量的颜色分量预测方法。
在本申请的一个实施例中，当第一图像块与第二图像块具有相同的分割树结构时，预测条件包括：目标区域中的第一颜色分量的非零系数的数量大于预设的数量阈值。
其中，第一颜色分量的非零系数指的是第一图像块中第一颜色分量的量化系数中取值不为零的量化系数，数量阈值可以是预设的默认值，或者也可以通过视频码流进行解析或者隐含导出得到。
举例而言，在色度块与亮度块具有相同的分割树结构时，即合并树模式下，如果与色度块对应的亮度块中非零系数的数量，大于或等于指定阈值，则允许使用亮度分量预测色度分量，否则禁止使用亮度分量预测色度分量。
在本申请的另一个实施例中，当第一图像块与第二图像块具有不同的分割树结构时，预测条件包括以下条件中的一种或多种：目标区域中的第一颜色分量的非零系数的数量，大于预设的数量阈值；目标区域对应的多个第一图像块的第一颜色分量的非零系数的数量之和，大于预设的数量阈值。
在色度块与亮度块具有不同的分割树结构时，即分离树模式下，如果色度块对应的亮度区域(即与色度块相对应的至少一个亮度块中的部分区域或全部区域)中非零系数的数量，大于等于指定阈值，则允许使用亮度分量预测色度分量，否则禁止使用亮度分量预测色度分量。
又例如，在色度块与亮度块具有不同的分割树结构时，即分离树模式下，如果色度块对应的亮度区域所在的所有亮度块中非零系数的数量总和，大于或等于指定阈值，则允许使用亮度分量预测色度分量，否则禁止使用亮度分量预测色度分量。
在本申请的一个实施例中，预测条件包括以下条件中的一种或多种：第一图像块与所述第二图像块具有相同的分割树结构，并且目标区域中的第一颜色分量的非零系数的数量大于预设的数量阈值；或者，第一图像块与第二图像块具有不同的分割树结构。
举例而言，对于亮度色度合并树的情况，如果色度块对应的亮度块中的非零系数的数量大于或等于3，则允许使用跨分量预测，否则禁止使用跨分量预测。对于亮度色度分离树的情况，不受对应亮度块中非零系数的数量限制，均可以使用跨分量预测。
在本申请的一个实施例中，预测条件还可以包括以下五种条件中的至少一种：
第一种预测条件：采用预设的预测模式，对第二图像块进行预测，预测模式包括帧内预测模式、帧间预测模式和帧内块复制预测模式中的一种或多种。
第二种预测条件：与第二图像块相对应的索引，具有指定的索引值，该索引用于指示是否执行：对第一颜色分量的重建值进行映射，得到第二图像块的第二颜色分量的预测值。
例如，针对一个第二图像块，可以在序列头/图像头/片头/最大编码块中解码一个索引，该索引用于指示当前序列/图像/片/最大编码块中的所有图像块是否执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。
第三种预测条件：目标区域的图像特征，满足预设的第一特征条件。
在本申请的一个实施例中，第一特征条件包括：目标区域中的第一颜色分量的重建参数处于预设的数值范围内。
在本申请的一个实施例中，预设的数值范围包括大于零或者小于零。
例如，重建参数指上述重建的残差信号，简称重建残差。针对一个第二图像块，只有当与之对应的目标区域中的第一颜色分量的重建残差大于零或者小于零时，才对第二图像块执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。而当目标区域中的第一颜色分量的重建残差等于零时，可以跳过依据第一颜色分量预测第二颜色分量的跨分量预测步骤。
第四种颜色分量预测条件：第二图像块的图像特征，满足预设的第二特征条件。
在本申请的一个实施例中，第二特征条件可以包括：第二图像块的图像尺寸，处于预设的尺寸范围内；或者，第二图像块在当前视频帧中的位置，处于指定的区域范围内。
在一种可选的实施方式中，可以根据第二图像块的图像尺寸大小，是否符合预设的范围限制，指示第二图像块是否执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。
在一种可选的实施方式中，可以根据第二图像块的位置，是否符合预设的范围限制，指示第二图像块是否执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。举例而言，如果第二图像块位于当前视频帧的左上角，则不执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。
第五种颜色分量预测条件：与第二图像块相对应的参考区域的图像特征，满足预设的第三特征条件，该参考区域包括与第二图像块相邻的至少一个图像区域。
在本申请的一个实施例中，第三特征条件包括：与第二图像块相对应的参考区域的区域面积，大于指定的面积阈值；或者，与第二图像块相对应的参考区域中指定采样位置的像素数量，大于预设的数量阈值。其中，预设的数量阈值可以是预测模型中的模型参数的数量。
在一种可选的实施方式中，只有当参考区域的区域面积大于指定的面积阈值时，才对第二图像块执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。
在一种可选的实施方式中，只有在参考区域中指定采样位置的像素数量(即第一颜色分量与第二颜色分量的匹配对中包括的像素数量)大于预测模型中的模型参数的数量时，才对第二图像块执行依据第一颜色分量预测第二颜色分量的跨分量预测方法。
在本申请的一个实施例中，在根据映射关系对目标区域的第一颜色分量的重建值进行映射处理之前，可以先对第一颜色分量进行下采样，获得与第二颜色分量能够形成匹配对的第一颜色分量。
以图5所示的YUV格式的图像块为例，当第一颜色分量为亮度分量Y，而第二颜色分量为色度分量U或者色度分量V时，按照420采样格式进行颜色分量采样后得到的亮度块和色度块具有不同的块尺寸。在水平方向和竖直方向上，亮度分量的分辨率都是色度分量的二倍，此时为了将亮度分量和色度分量组成匹配对，可以对亮度分量进行下采样。
在本申请的一个实施例中，对第一颜色分量进行下采样的方法可以包括：根据第二图像块的第二颜色分量的像素点位置，确定具有指定窗口尺寸的采样窗口；在采样窗口中，对目标区域的第一颜色分量的重建值进行下采样，得到与像素点位置相匹配的第一颜色分量的重建值。在对目标区域的第一颜色分量的重建值进行映射时，可以对匹配的第一颜色分量的重建值进行映射，得到第二图像块的第二颜色分量的预测值。
在本申请的一个实施例中，根据第二颜色分量的像素点位置确定的采样窗口，用于覆盖最近邻的四个第一颜色分量以及位于左侧的次近邻的两个第一颜色分量。
图16示出了本申请一个实施例中对第一颜色分量进行下采样的采样窗口示意图。
如图16所示，图中以阵列形式分布的方块表示第一颜色分量，包括16011-16016，而在阵列内分布的五角星表示第二颜色分量，包括16021和16022。本申请实施例中，可以根据第二颜色分量16021所在的像素点位置，确定与之对应的具有指定窗口尺寸的采样窗口1603，其中，采样窗口1603能够覆盖与第二颜色分量16021最近邻的四个第一颜色分量16011-16014以及位于左侧的次近邻的两个第一颜色分量16015、16016。
在本申请的一个实施例中，将目标区域的第一颜色分量的重建值在采样窗口中进行下采样，得到与像素点位置相匹配的第一颜色分量的重建值，包括：在采样窗口中，获取目标区域的第一颜色分量的重建值与像素点位置之间的对应关系；根据对应关系，对第一颜色分量的重建值进行加权运算，得到与像素点位置相匹配的第一颜色分量的重建值。
举例而言，本申请实施例可以按照如下公式对第一颜色分量的重建值进行下采样：
L’(x,y)＝(L(x-1,y)+2L(x,y)+L(x+1,y)+L(x-1,y+1)+2L(x,y+1)+L(x+1,y+1))/8
其中，L和L′分别表示原始的第一颜色分量的重建值和下采样后的第一颜色分量的重建值。x和y分别表示第一颜色分量的水平位置坐标和竖直位置坐标。以图16所示的采样窗口为例，位于第二颜色分量16021左侧的最近邻的两个第一颜色分量16011、16013，其坐标为(x,y)和(x,y+1)，可以分配权重系数为1/4；除此之外的另外四个第一颜色分量16012、16014-16016，其坐标为(x-1,y)、x-1,y+1)、(x+1,y)以及(x+1,y+1)，可以分配权重系数为1/8。
在本申请的一个实施例中，当位于左侧的次近邻的两个第一颜色分量16015、16016不可用，其坐标为(x-1,y)、(x-1,y+1)的两个第一颜色分量不可用时，可以使用位于左侧的最近邻的两个第一颜色分量16011、16013代替，即使用坐标为(x,y)或者(x,y+1)的两个第一颜色分量代替。
在本申请的一个实施例中，在对预测模型进行训练时，当选择对目标区域的第一颜色分量进行下采样时，可以同步地对参考区域中的至少一个图像区域内对作为模型训练样本的第一颜色分量进行下采样。即，在参考区域中的至少一个图像区域内，根据第二图像块的第二颜色分量的像素点位置，确定具有指定窗口尺寸的采样窗口，参考区域包括与第二图像块相邻的至少一个图像区域；在采样窗口中，对至少一个图像区域的第一颜色分量的重建值进行下采样，得到与像素点位置相匹配的第一颜色分量的重建值。
在本申请的一个实施例中，可以选择对第一颜色分量进行下采样，或者也可以不进行下采样，而选择与第二颜色分量位置相同或相应的第一颜色分量执行颜色分量预测。以图16为例，可以选择与第二颜色分量1602相对应的位置坐标为(x，y)的第一颜色分量1601执行颜色分量预测。
在本申请的一个实施例中，当第一颜色分量与第二颜色分量的采样格式相同时，在通过预测模型对第一颜色分量进行颜色分量预测之前，可以先对第一颜色分量进行滤波处理，减少第一颜色分量中的冗余信息或者干扰信息，从而进一步提高第二颜色分量的预测准确性，即提高第二图像块的视频编解码准确性。
在此基础上，在拟合预测模型时，可以同时选用多个滤波器对第一颜色分量的样本数据进行滤波处理，进而可以选取不同的第一颜色分量样本训练拟合预测模型，提高预测模型的拟合训练效果。
在本申请的一个实施例中，在对第一颜色分量进行下采样或者滤波处理后，可以根据需要对第一颜色分量的像素点位置进行边界扩展，从而获得用于指定像素点的邻域像素点。
需要说明的是，以上实施例中介绍了多种不同的跨分量预测方案，例如参考区域包括多种不同候选区域模板，预测模型也包括多种不同的候选预测模型，将多种不同的候选区域模板与多种不同的候选预测模型进行组合，可以得到多种不同的颜色分量预测方案。
对于不同的图像块，可以选择使用相同的跨分量预测方案或者使用不同的跨分量预测方案。
在本申请的一个实施例中，例如具有不同块尺寸的图像块可以选用不同的颜色分量预测方案(候选区域模板、候选预测模型)。
在本申请的一个实施例中，可以在数据码流中通过显式索引或者隐式索引指示当前块所使用的颜色分量预测方案，例如可以通过一个索引或者多个索引的组合，指示当前块使用的候选区域模板和/或候选预测模型。
在本申请的一个实施例中，可以对一个当前块使用多种不同的颜色分量预测方案，然后将多种方案的预测结果进行加权运算，得到最终的颜色分量预测结果。
在本申请的一个实施例中，最终的预测图像可以由以上实施例提供的跨分量预测结果和普通的帧内预测模式、帧间预测模式或者帧内块复制预测模式的预测结果进行加权得到。
在本申请的一个实施例中，根据预设的预测模式，例如，对第二图像块进行帧内预测、帧间预测或者帧内块复制预测，得到第二图像块的第二颜色分量的初始预测值；在根据映射关系对目标区域的第一颜色分量的重建值进行映射处理，得到第二图像块的第二颜色分量的预测值之后，根据预设的权重系数，对初始预测值与第二颜色分量的预测值进行加权运算，得到第二颜色分量的更新后的预测值。
举例而言，本申请实施例可以按照如下公式确定第二图像块的第二颜色分量的最终预测结果pred。
pred＝w*pred1+(1-w)*pred2
其中，pred1表示根据以上任意实施例提供的方案依据第一颜色分量预测得到的第二颜色分量的预测值，pred2表示根据帧内预测模式、帧间预测模式或者帧内块复制预测模式预测得到的第二颜色分量的初始预测值，w表示对两种预测结果进行加权运算的权重系数。
以下介绍使用本申请实施例在一个应用场景中的实现过程。
在该应用场景中，使用帧间预测模式或者帧内块复制预测模式对亮度块对应的亮度分量Y进行预测，得到亮度块的亮度分量重建值(即上述目标区域的第一颜色分量的重建值)，然后，使用亮度分量Y分别预测色度分量U和色度分量V。
结合图9所示，采用的预测模型如下所示：
该应用场景中使用了与当前块相对应的参考区域训练预测模型，即选取参考区域中的像素点作为训练预测模型的样本像素点。相应地，训练样本对包括参考区域中的样本像素点的亮度分量重建值和色度分量重建值。
其中，参考区域可以选取图10所示的与当前块最近邻和次近邻的多个图像区域。即，选取位于当前块上方和左侧的两个最近邻区域以及位于当前块左上方、左下方和右上方的三个次近邻区域。
之后，可以对亮度块的亮度分量重建值和参考区域的亮度分量重建值进行下采样，例如可以使用图16所示的采样窗口。下采样过程如下所示：
L’(x,y)＝(L(x-1,y)+2L(x,y)+L(x+1,y)+L(x-1,y+1)+2L(x,y+1)+L(x+1,y+1))/8
在完成亮度分量重建值的下采样后，在参考区域中选取位置坐标在水平方向和竖直方向均不为奇数的像素作为样本像素点，构建包括亮度分量重建值与色度分量重建值的样本对。在其他可选的实现方式中，也可以在参考区域中选取在水平方向和竖直方向上至少一个位置坐标为偶数的像素作为样本像素点。
将样本对输入至预测模型，建立方程后求解预测模型中的各个模型参数pi，i＝0,…,6，分别得到用于根据亮度分量Y预测色度分量U的第一预测模型以及用于根据亮度分量Y预测色度分量V的第二预测模型。
对亮度块中的亮度重建值进行下采样后，分别输入至拟合训练完成的第一预测模型和第二预测模型，获得色度分量U和色度分量V的预测结果。
如果当前预测模式是帧间预测模式或者帧内块复制预测模式，则按照如下公式分别将色度分量U和色度分量V的预测结果与初始预测结果进行加权运算，得到色度分量U和色度分量V的最终预测结果pred。初始预测结果包括帧间预测的预测结果或者帧内块复制的预测结果。
pred＝w*pred1+(1-w)*pred2
其中，pred1表示根据依据亮度分量Y预测得到的色度分量U或色度分量V的预测值，pred2表示根据帧间预测模式或者帧内块复制预测模式预测得到的色度分量U或色度分量V的初始预测值，w表示对两种预测结果进行加权运算的权重系数，该权重系数例如可以取值为0.75。
在该应用场景中，允许亮度色度合并树或者亮度色度分离树使用本申请实施例提供的跨分量预测方案。对于亮度色度合并树的情况，色度对应亮度块的非零系数数量大于等于3允许使用跨分量预测，否则禁止使用跨分量预测。对于亮度色度分离树的情况，不受对应亮度块的非零系数数量限制。
基于以上各个实施例以及应用场景的介绍可知，本申请实施例提出了一种跨分量的颜色分量预测方法，通过利用邻近区域或者预测图像与重建图像的分量间映射关系的相似性，基于邻近已重建像素或者当前块的预测图像生成跨分量预测模型，然后再利用跨分量预测模型生成当前图像块的目标预测图像。相比于常规的编码预测方案，本申请实施例可以进一步利用当前块的已重建像素提升颜色分量的预测精度，从而提升了编解码效率。而且，本申请实施例扩大了跨分量预测的使用范围，即亮度色度分离树模式也可以采用跨分量的颜色分量预测方案。另外，当色度块对应位置的非零亮度系数数量较少时不允许使用跨分量预测，以此达到节省码率、提升编解码效率的效果。
应当注意，尽管在附图中以特定顺序描述了本申请中方法的各个步骤，但是，这并非要求或者暗示必须按照该特定顺序来执行这些步骤，或是必须执行全部所示的步骤才能实现期望的结果。附加的或备选的，可以省略某些步骤，将多个步骤合并为一个步骤执行，以及/或者将一个步骤分解为多个步骤执行等。
以下介绍本申请的装置实施例，可以用于执行本申请上述实施例中的视频编解码方法。
图17示意性地示出了本申请实施例提供的图像处理装置的结构框图。如图17所示，图像处理装置1700可以包括：
分割模块1710，被配置为对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，所述第一图像块是具有第一颜色分量的编码单元，所述第二图像块是具有第二颜色分量的编码单元；
确定模块1720，被配置为根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域；
预测模块1730，被配置为根据预设的预测模式，预测得到所述目标区域的所述第一颜色分量的重建值；及，
映射模块1740，被配置为响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
在本申请的一个实施例中，分割模块1710，被配置为基于单个分割树结构，对所述当前视频帧进行分割，得到所述至少一个第一图像块和所述第二图像块；
确定模块1720，被配置为基于所述位置，将所述至少一个第一图像块中与所述第二图像块相对应的一个第一图像块的全部区域，确定为所述目标区域。
在本申请的一个实施例中，分割模块1710，被配置为基于第一分割树结构，对所述当前视频帧进行分割，得到所述至少一个第一图像块；基于第二分割树结构，对所述当前视频帧进行分割，得到所述第二图像块；
确定模块1720，被配置为基于所述位置，将所述至少一个第一图像块中与所述第二图像块相对应的部分区域或者全部区域，确定为所述目标区域。
在本申请的一个实施例中，所述预测条件包括：所述目标区域的所述第一颜色分量的非零系数的数量，大于预设的数量阈值。
在本申请的一个实施例中，所述预测条件包括以下条件中的一种或多种：
所述第一分割树结构与所述第二分割树结构不同；
所述目标区域的所述第一颜色分量的非零系数的数量，大于预设的数量阈值；
所述目标区域对应的多个第一图像块的所述第一颜色分量的非零系数的数量之和，大于预设的数量阈值。
在本申请的一个实施例中，所述预测条件包括以下条件中的一种或多种：
与所述第二图像块相对应的索引，具有指定的索引值，所述索引用于指示是否执行：对所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值；
所述目标区域的图像特征，满足预设的第一特征条件；
所述第二图像块的图像特征，满足预设的第二特征条件；
与所述第二图像块相对应的参考区域的图像特征，满足预设的第三特征条件，所述参考区域包括与所述第二图像块相邻的至少一个图像区域。
在本申请的一个实施例中，所述第一特征条件包括：所述目标区域的所述第一颜色分量的重建参数，处于预设的数值范围内。
在本申请的一个实施例中，所述第二特征条件包括以下条件中的一种或多种：
所述第二图像块的图像尺寸，处于预设的尺寸范围内；
所述第二图像块在所述当前视频帧中的所述位置，处于指定的区域范围内。
所述第三特征条件包括以下条件中的一种或多种：
与所述第二图像块相对应的所述参考区域的区域面积，大于预设的面积阈值；
与所述第二图像块相对应的所述参考区域中指定采样位置的像素数量，大于预设的数量阈值。
在本申请的一个实施例中，所述映射关系由预测模型表示，所述数量阈值是所述预测模型中的模型参数的数量。
在本申请的一个实施例中，所述预测装置1700还包括：
加权模块1750，被配置为根据预设的预测模式，预测得到所述第二图像块的所述第二颜色分量的初始预测值，所述预测模式包括帧内预测模式、帧间预测模式或者帧内块复制预测模式；
根据预设的权重系数，对所述初始预测值与所述第二颜色分量的预测值进行加权运算，得到所述第二颜色分量的更新后的预测值。
在本申请的一个实施例中，所述映射关系由预测模型表示，所述预测装置1700还包括：
初始模型获取模块1760，被配置为获取初始预测模型，所述初始预测模型的输入项包括指定像素点的所述第一颜色分量，所述初始预测模型的输出项为所述指定像素点的所述第二颜色分量，所述初始预测模型包括多种候选模型中的至少一种；
样本对选取模块1770，被配置为获取训练样本对，所述训练样本对包括以下样本对中的一种或多种：
位于所述目标区域内的样本像素点的所述第一颜色分量的预测值和位于所述第二图像块内的样本像素点的所述第二颜色分量的预测值；
位于参考区域内的样本像素点的所述第一颜色分量的重建值和所述第二颜色分量的重建值，所述参考区域包括至少一个与所述第二图像块相邻的图像区域；
模型拟合模块1780，被配置为根据所述训练样本对，对所述初始预测模型进行参数拟合，得到所述预测模型。
在本申请的一个实施例中，位于所述目标区域内的样本像素点的所述第一颜色分量的预测值和位于所述第二图像块内的样本像素点的所述第二颜色分量的预测值，是根据预设的预测模式预测得到的，所述预测模式包括帧内预测模式、帧间预测模式或者帧内块复制预测模式。
在本申请的一个实施例中，所述初始预测模型的输入项还包括至少一个邻域像素点的所述第一颜色分量，所述邻域像素点是与所述指定像素点最近邻或次近邻的像素点。
在本申请的一个实施例中，所述初始预测模型包括至少一个具有独立的加权参数的组合项，所述组合项以至少两个像素点的所述第一颜色分量作为输入项。
在本申请的一个实施例中，所述初始预测模型包括至少两个具有不同阶数的所述组合项，所述阶数是所述组合项中输入项的最高次幂。
在本申请的一个实施例中，所述参考区域由至少一个最近邻区域或次近邻区域组合而成，所述最近邻区域包括位于所述第二图像块上方或左侧的具有指定图像尺寸的图像区域，所述次近邻区域包括位于所述第二图像块左上方、左下方或右上方的具有指定图像尺寸的图像区域。
在本申请的一个实施例中，所述第二图像块右上方的次近邻区域在水平方向上与所述第二图像块具有相同的图像尺寸，所述第二图像块右上方的次近邻区域在竖直方向上具有指定图像尺寸；
所述第二图像块左下方的次近邻区域在竖直方向上与所述第二图像块具有相同的图像尺寸，所述第二图像块右上方的次近邻区域在水平方向上具有指定图像尺寸；
所述第二图像块左上方的次近邻区域在水平方向上和竖直方向上均具有指定图像尺寸。
在本申请的一个实施例中，所述参考区域包括全量区域组合、左侧区域组合与上方区域组合中的至少一种；
其中，所述全量区域组合包括位于所述第二图像块左侧和上方的最近邻区域以及位于所述第二图像块左上方、左下方和右上方的次近邻区域；
所述左侧区域组合包括位于所述第二图像块左侧的最近邻区域以及位于所述第二图像块左下方的次近邻区域；
所述上方区域组合包括位于所述第二图像块上方的最近邻区域以及位于所述第二图像块右上方的次近邻区域。
在本申请的一个实施例中，所述预测装置1700还包括：
区域调整模块1790，被配置为获取组成所述参考区域的至少一个子区域的可用性信息；根据所述可用性信息，调整所述参考区域的区域范围。
在本申请的一个实施例中，所述区域调整模块1790被配置为从所述参考区域中移除处于不可用状态的子区域；当所述参考区域中的所有子区域均处于不可用状态时，将所述参考区域配置为不可用状态。
在本申请的一个实施例中，所述样本对获取模块1770被配置为：获取所述第二图像块的像素点采样模式，所述像素点采样模式包括全量像素点采样或者部分像素点采样；
当所述第二图像块的像素点采样模式为所述全量像素点采样时，所述位于所述第二图像块内的样本像素点是指所述第二图像块内的所有像素点，所述位于所述参考区域内的样本像素点是指所述参考区域中的至少一个图像区域内的所有像素点；
当所述第二图像块的像素点采样模式为所述部分像素点采样时，所述位于所述第二图像块内的样本像素点是指处于指定采样位置的部分像素点，所述位于所述参考区域内的样本像素点是指所述参考区域中的至少一个图像区域内处于指定采样位置的部分像素点。
在本申请的一个实施例中，所述指定采样位置满足以下条件中的至少一种：
所述指定采样位置的坐标，满足预设的坐标数值条件；
沿预设的像素点扫描方向，选取所述指定采样位置；
所述指定采样位置的像素点的所述第一颜色分量的重建值，处于预设的数值范围内。
在本申请的一个实施例中，所述坐标数值条件包括：
水平位置坐标与竖直位置坐标中的至少一个为偶数；或者，
水平位置坐标与竖直位置坐标中的至少一个为奇数。
在本申请的一个实施例中，所述预测装置1700还包括：
第一下采样模块1791，被配置为根据所述第二图像块的所述第二颜色分量的像素点位置，确定具有指定窗口尺寸的采样窗口；在所述采样窗口中，对所述目标区域的所述第一颜色分量的重建值进行下采样，得到与所述像素点位置相匹配的所述第一颜色分量的重建值；
映射模块1740，被配置为对匹配的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
在本申请的一个实施例中，所述第一下采样模块1791被配置为：在所述采样窗口中，获取所述目标区域的第一颜色分量的重建值与所述像素点位置之间的对应关系；
根据所述对应关系，对所述目标区域的所述第一颜色分量的重建值进行加权运算，得到与所述像素点位置相匹配的所述第一颜色分量的重建值。
在本申请的一个实施例中，所述预测装置1700还包括：
第二下采样模块1792，被配置为在与所述第二图像块相对应的参考区域中的至少一个图像区域内，根据所述第二图像块的所述第二颜色分量的像素点位置，确定具有指定窗口尺寸的采样窗口；在所述采样窗口中，对所述至少一个图像区域的所述第一颜色分量的重建值进行下采样，得到与所述像素点位置相匹配的所述第一颜色分量的重建值。
本申请各实施例中提供的视频解码装置和视频编码装置的具体细节已经在对应的方法实施例中进行了详细的描述，此处不再赘述。
图18示意性地示出了用于实现本申请实施例的电子设备的结构框图。
需要说明的是，图18示出的电子设备的1800仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。
如图18所示，电子设备1800包括中央处理器1801(Central Processing Unit，CPU)，其可以根据存储在只读存储器1802(Read-Only Memory，ROM)中的程序或者从存储部分1808加载到随机访问存储器1803(Random Access Memory，RAM)中的程序而执行各种适当的动作和处理。在随机访问存储器1803中，还存储有系统操作所需的各种程序和数据。中央处理器1801、在只读存储器1802以及随机访问存储器1803通过总线1804彼此相连。输入/输出接口1805(Input/Output接口，即I/O接口)也连接至总线1804。
以下部件连接至输入/输出接口1805：包括键盘、鼠标等的输入部分1806；包括诸如阴极射线管(Cathode Ray Tube，CRT)、液晶显示器(Liquid Crystal Display，LCD)等以及扬声器等的输出部分1807；包括硬盘等的存储部分1808；以及包括诸如局域网卡、调制解调器等的网络接口卡的通信部分1809。通信部分1809经由诸如因特网的网络执行通信处理。驱动器1810也根据需要连接至输入/输出接口1805。可拆卸介质1811，诸如磁盘、光盘、磁光盘、半导体存储器等等，根据需要安装在驱动器1810上，以便于从其上读出的计算机程序根据需要被安装入存储部分1808。
特别地，根据本申请的实施例，各个方法流程图中所描述的过程可以被实现为计算机软件程序。例如，本申请的实施例包括一种计算机程序产品，其包括承载在计算机可读存储介质上的计算机程序，该计算机程序包含用于执行流程图所示的方法的程序代码。在这样的实施例中，该计算机程序可以通过通信部分1809从网络上被下载和安装，和/或从可拆卸介质1811被安装。在该计算机程序被中央处理器1801执行时，执行本申请的系统中限定的各种功能。
需要说明的是，本申请实施例所示的计算机可读存储介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有至少一个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(Erasable Programmable Read Only Memory，EPROM)、闪存、光纤、便携式紧凑磁盘只读存储器(Compact Disc Read-Only Memory，CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本申请中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。而在本申请中，计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读信号介质还可以是计算机可读存储介质以外的任何计算机可读存储介质，该计算机可读存储介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。计算机可读存储介质上包含的程序代码可以用任何适当的介质传输，包括但不限于：无线、有线等等，或者上述的任意合适的组合。
附图中的流程图和框图，图示了按照本申请各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，上述模块、程序段、或代码的一部分包含至少一个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图或流程图中的每个方框、以及框图或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
应当注意，尽管在上文详细描述中提及了用于动作执行的设备的若干模块或者单元，但是这种划分并非强制性的。实际上，根据本申请的实施方式，上文描述的两个或更多模块或者单元的特征和功能可以在一个模块或者单元中具体化。反之，上文描述的一个模块或者单元的特征和功能可以进一步划分为由多个模块或者单元来具体化。
通过以上的实施方式的描述，本领域的技术人员易于理解，这里描述的示例实施方式可以通过软件实现，也可以通过软件结合必要的硬件的方式来实现。因此，根据本申请实施方式的技术方案可以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM，U盘，移动硬盘等)中或网络上，包括若干指令以使得一台计算设备(可以是个人计算机、服务器、触控终端、或者网络设备等)执行根据本申请实施方式的方法。
本领域技术人员在考虑说明书及实践这里公开的发明后，将容易想到本申请的其它实施方案。本申请旨在涵盖本申请的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。
应当理解的是，本申请并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本申请的范围仅由所附的权利要求来限制。

Claims (31)

  1. 一种图像处理方法，由电子设备执行，包括：
    对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，所述第一图像块是具有第一颜色分量的编码单元，所述第二图像块是具有第二颜色分量的编码单元；
    根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域；
    根据预设的预测模式，预测得到所述目标区域的所述第一颜色分量的重建值；及，
    响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
  2. 根据权利要求1所述的方法，其中，所述对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，包括：
    基于单个分割树结构，对所述当前视频帧进行分割，得到所述至少一个第一图像块和所述第二图像块；
    所述根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域，包括：
    基于所述位置，将所述至少一个第一图像块中与所述第二图像块相对应的一个第一图像块的全部区域，确定为所述目标区域。
  3. 根据权利要求1所述的方法，其中，所述对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，包括：
    基于第一分割树结构，对所述当前视频帧进行分割，得到所述至少一个第一图像块；
    基于第二分割树结构，对所述当前视频帧进行分割，得到所述第二图像块；
    所述根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域，包括：
    基于所述位置，将所述至少一个第一图像块中与所述第二图像块相对应的部分区域或者全部区域，确定为所述目标区域。
  4. 根据权利要求2所述的方法，其中，所述预测条件包括：所述目标区域的所述第一颜色分量的非零系数的数量，大于预设的数量阈值。
  5. 根据权利要求3所述的方法，其中，所述预测条件包括以下条件中的一种或多种：
    所述第一分割树结构与所述第二分割树结构不同；
    所述目标区域的所述第一颜色分量的非零系数的数量，大于预设的数量阈值；
    所述目标区域对应的多个第一图像块的所述第一颜色分量的非零系数的数量之和，大于预设的数量阈值。
  6. 根据权利要求1-3中任一项所述的方法，其中，所述预测条件包括以下条件中的一种或多种：
    与所述第二图像块相对应的索引，具有指定的索引值，所述索引用于指示是否执行：对所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值；
    所述目标区域的图像特征，满足预设的第一特征条件；
    所述第二图像块的图像特征，满足预设的第二特征条件；
    与所述第二图像块相对应的参考区域的图像特征，满足预设的第三特征条件，所述参考区域包括与所述第二图像块相邻的至少一个图像区域。
  7. 根据权利要求6所述的方法，其中，所述第一特征条件包括：所述目标区域的所述第一颜色分量的重建参数，处于预设的数值范围内。
  8. 根据权利要求6或者7所述的方法，其中，所述第二特征条件包括以下条件中的一种或多种：
    所述第二图像块的图像尺寸，处于预设的尺寸范围内；
    所述第二图像块在所述当前视频帧中的所述位置，处于指定的区域范围内。
  9. 根据权利要求6-8中任一项所述的方法，其中，所述第三特征条件包括以下条件中的一种或多种：
    与所述第二图像块相对应的所述参考区域的区域面积，大于预设的面积阈值；
    与所述第二图像块相对应的所述参考区域中指定采样位置的像素数量，大于预设的数量阈值。
  10. 根据权利要求9所述的方法，其中，所述映射关系由预测模型表示，所述数量阈值是所述预测模型中的模型参数的数量。
  11. 根据权利要求1-10中任一项所述的方法，还包括：
    根据预设的预测模式，预测得到所述第二图像块的所述第二颜色分量的初始预测值，所述预测模式包括帧内预测模式、帧间预测模式或者帧内块复制预测模式；
    根据预设的权重系数，对所述初始预测值与所述第二颜色分量的预测值进行加权运算，得到所述第二颜色分量的更新后的预测值。
  12. 根据权利要求1-11中任一项所述的方法，其中，所述映射关系由预测模型表示，所述方法还包括：
    获取所述预测模型，包括：
    获取初始预测模型，所述初始预测模型的输入项包括指定像素点的所述第一颜色分量，所述初始预测模型的输出项为所述指定像素点的所述第二颜色分量，所述初始预测模型包括多种候选模型中的至少一种；
    获取训练样本对，所述训练样本对包括以下样本对中的一种或多种：
    位于所述目标区域内的样本像素点的所述第一颜色分量的预测值和位于所述第二图像块内的样本像素点的所述第二颜色分量的预测值；
    位于参考区域内的样本像素点的所述第一颜色分量的重建值和所述第二颜色分量的重建值，所述参考区域包括至少一个与所述第二图像块相邻的图像区域；
    根据所述训练样本对，对所述初始预测模型进行参数拟合，得到所述预测模型。
  13. 根据权利要求12所述的方法，其中，位于所述目标区域内的样本像素点的所述第一颜色分量的预测值和位于所述第二图像块内的样本像素点的所述第二颜色分量的预测值，是根据预设的预测模式预测得到的，所述预测模式包括帧内预测模式、帧间预测模式或者帧内块复制预测模式。
  14. 根据权利要求12或者13所述的方法，其中，所述初始预测模型的输入项还包括至少一个邻域像素点的所述第一颜色分量，所述邻域像素点是与所述指定像素点最近邻或次近邻的像素点。
  15. 根据权利要求14所述的方法，其中，所述初始预测模型包括至少一个具有独立的加权参数的组合项，所述组合项以至少两个像素点的所述第一颜色分量作为输入项。
  16. 根据权利要求15所述的方法，其中，所述初始预测模型包括至少两个具有不同阶数的所述组合项，所述阶数是所述组合项中输入项的最高次幂。
  17. 根据权利要求12-16中任一项所述的方法，其中，所述参考区域由至少一个最近邻区域或次近邻区域组合而成，所述最近邻区域包括位于所述第二图像块上方或左侧的具有指定图像尺寸的图像区域，所述次近邻区域包括位于所述第二图像块左上方、左下方或右上方的具有指定图像尺寸的图像区域。
  18. 根据权利要求17所述的方法，其中，所述第二图像块右上方的次近邻区域在水平方向上与所述第二图像块具有相同的图像尺寸，所述第二图像块右上方的次近邻区域在竖直方向上具有指定图像尺寸；
    所述第二图像块左下方的次近邻区域在竖直方向上与所述第二图像块具有相同的图像尺寸，所述第二图像块右上方的次近邻区域在水平方向上具有指定图像尺寸；
    所述第二图像块左上方的次近邻区域在水平方向上和竖直方向上均具有指定图像尺寸。
  19. 根据权利要求17所述的方法，其中，所述参考区域包括全量区域组合、左侧区域组合与上方区域组合中的至少一种；
    其中，所述全量区域组合包括位于所述第二图像块左侧和上方的最近邻区域以及位于所述第二图像块左上方、左下方和右上方的次近邻区域；
    所述左侧区域组合包括位于所述第二图像块左侧的最近邻区域以及位于所述第二图像块左下方的次近邻区域；
    所述上方区域组合包括位于所述第二图像块上方的最近邻区域以及位于所述第二图像块右上方的次近邻区域。
  20. 根据权利要求17-19中任一项所述的方法，还包括：
    获取组成所述参考区域的至少一个子区域的可用性信息；
    根据所述可用性信息，调整所述参考区域的区域范围。
  21. 根据权利要求20所述的方法，其中，所述根据所述可用性信息，调整所述参考区域的区域范围，包括：
    从所述参考区域中移除处于不可用状态的子区域；
    当所述参考区域中的所有子区域均处于不可用状态时，将所述参考区域配置为不可用状态。
  22. 根据权利要求12-21中任一项所述的方法，其中，所述获取训练样本对，包括：
    获取所述第二图像块的像素点采样模式，所述像素点采样模式包括全量像素点采样或者部分像素点采样；
    当所述第二图像块的像素点采样模式为所述全量像素点采样时，所述位于所述第二图像块内的样本像素点是指所述第二图像块内的所有像素点，所述位于所述参考区域内的样本像素点是指所述参考区域中的至少一个图像区域内的所有像素点；
    当所述第二图像块的像素点采样模式为所述部分像素点采样时，所述位于所述第二图像块内的样本像素点是指处于指定采样位置的部分像素点，所述位于所述参考区域内的样本像素点是指所述参考区域中的至少一个图像区域内处于指定采样位置的部分像素点。
  23. 根据权利要求22所述的方法，其中，所述指定采样位置满足以下条件中的至少一种：
    所述指定采样位置的坐标，满足预设的坐标数值条件；
    沿预设的像素点扫描方向，选取所述指定采样位置；
    所述指定采样位置的像素点的所述第一颜色分量的重建值，处于预设的数值范围内。
  24. 根据权利要求23所述的方法，其中，所述坐标数值条件包括：
    水平位置坐标与竖直位置坐标中的至少一个为偶数；或者，
    水平位置坐标与竖直位置坐标中的至少一个为奇数。
  25. 根据权利要求1-24中任一项所述的方法，还包括：
    根据所述第二图像块的所述第二颜色分量的像素点位置，确定具有指定窗口尺寸的采样窗口；
    在所述采样窗口中，对所述目标区域的所述第一颜色分量的重建值进行下采样，得到与所述像素点位置相匹配的所述第一颜色分量的重建值；
    所述响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值，包括：
    对匹配的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
  26. 根据权利要求25所述的方法，其中，所述将所述目标区域中的所述第一颜色分量的重建值在所述采样窗口中进行下采样，得到与所述像素点位置相匹配的所述第一颜色分量的重建值，包括：
    在所述采样窗口中，获取所述目标区域的第一颜色分量的重建值与所述像素点位置之间的对应关系；
    根据所述对应关系，对所述目标区域的所述第一颜色分量的重建值进行加权运算，得到与所述像素点位置相匹配的所述第一颜色分量的重建值。
  27. 根据权利要求25所述的方法，还包括：
    在与所述第二图像块相对应的参考区域中的至少一个图像区域内，根据所述第二图像块的所述第二颜色分量的像素点位置，确定具有指定窗口尺寸的采样窗口；
    在所述采样窗口中，对所述至少一个图像区域的所述第一颜色分量的重建值进行下采样，得到与所述像素点位置相匹配的所述第一颜色分量的重建值。
  28. 一种图像处理装置，包括：
    分割模块，被配置为对当前视频帧进行分割，得到至少一个第一图像块和第二图像块，其中，所述第一图像块是具有第一颜色分量的编码单元，所述第二图像块是具有第二颜色分量的编码单元；
    确定模块，被配置为根据所述第二图像块在所述当前视频帧中的位置，将所述至少一个第一图像块中与所述位置关联的编码区域，确定为目标区域；
    预测模块，被配置为根据预设的预测模式，预测得到所述目标区域的所述第一颜色分量的重建值；及，
    映射模块，被配置为响应于预设的预测条件被满足，根据所述第一颜色分量与所述第二颜色分量之间的映射关系，对所述目标区域的所述第一颜色分量的重建值进行映射，得到所述第二图像块的所述第二颜色分量的预测值。
  29. 一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1至27中任意一项所述的方法。
  30. 一种电子设备，包括：
    处理器；以及
    存储器，用于存储所述处理器的可执行指令；
    其中，所述处理器配置为执行所述可执行指令以实现如权利要求1至27中任意一项所述的方法。
  31. 一种计算机程序产品，包括计算机程序，所述计算机程序存储在计算机可读存储介质中；当电子设备的处理器从所述计算机可读存储介质读取所述计算机程序时，所述处理器执行所述计算机程序，使得所述电子设备执行权利要求1-27中任一项所述方法的步骤。
PCT/CN2025/077524 2024-03-14 2025-02-17 图像处理方法、装置、存储介质、电子设备及程序产品 Pending WO2025190039A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020267019671A KR20260113116A (ko) 2024-03-14 2025-02-17 이미지 프로세싱 방법 및 장치, 그리고 저장 매체, 전자 디바이스, 및 프로그램 제품

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410297228.5 2024-03-14
CN202410297228.5A CN120658875A (zh) 2024-03-14 2024-03-14 颜色分量的预测方法、装置、介质及电子设备

Publications (1)

Publication Number Publication Date
WO2025190039A1 true WO2025190039A1 (zh) 2025-09-18

Family

ID=96998646

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2025/077524 Pending WO2025190039A1 (zh) 2024-03-14 2025-02-17 图像处理方法、装置、存储介质、电子设备及程序产品

Country Status (3)

Country Link
KR (1) KR20260113116A (zh)
CN (1) CN120658875A (zh)
WO (1) WO2025190039A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220279170A1 (en) * 2019-11-21 2022-09-01 Nippon Hoso Kyokai Intra prediction device, image encoding device, image decoding device and program
WO2023224290A1 (ko) * 2022-05-18 2023-11-23 현대자동차주식회사 인트라 예측에서 성분간 관계 모델 유도를 위한 참조샘플 선택을 위한 방법 및 장치
WO2023239676A1 (en) * 2022-06-06 2023-12-14 Beijing Dajia Internet Information Technology Co., Ltd. Improved cross-component prediction for video coding
WO2024007165A1 (zh) * 2022-07-05 2024-01-11 Oppo广东移动通信有限公司 编解码方法、装置、编码设备、解码设备以及存储介质
WO2024034849A1 (ko) * 2022-08-09 2024-02-15 현대자동차주식회사 루마 성분 기반 크로마 성분 예측을 이용하는 비디오 코딩을 위한 방법 및 장치

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220279170A1 (en) * 2019-11-21 2022-09-01 Nippon Hoso Kyokai Intra prediction device, image encoding device, image decoding device and program
WO2023224290A1 (ko) * 2022-05-18 2023-11-23 현대자동차주식회사 인트라 예측에서 성분간 관계 모델 유도를 위한 참조샘플 선택을 위한 방법 및 장치
WO2023239676A1 (en) * 2022-06-06 2023-12-14 Beijing Dajia Internet Information Technology Co., Ltd. Improved cross-component prediction for video coding
WO2024007165A1 (zh) * 2022-07-05 2024-01-11 Oppo广东移动通信有限公司 编解码方法、装置、编码设备、解码设备以及存储介质
WO2024034849A1 (ko) * 2022-08-09 2024-02-15 현대자동차주식회사 루마 성분 기반 크로마 성분 예측을 이용하는 비디오 코딩을 위한 방법 및 장치

Also Published As

Publication number Publication date
KR20260113116A (ko) 2026-07-21
CN120658875A (zh) 2025-09-16

Similar Documents

Publication Publication Date Title
US12556696B2 (en) Video coding method and apparatus, medium, and electronic device
JP7717267B2 (ja) ループフィルタリング方法、ビデオ復号方法、ビデオ符号化方法、ループフィルタリング装置、ビデオ復号装置、ビデオ符号化装置、電子機器及びコンピュータプログラム
CN117136540A (zh) 残差编码方法及设备、视频编码方法及设备、存储介质
WO2023173809A1 (zh) 视频编解码方法、装置、存储介质、电子设备及计算机程序产品
US20250071306A1 (en) Video coding/decoding method and apparatus, computer-readable medium, and electronic device
CN114979655B (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN118984390B (zh) 用于视频编解码的基于历史的莱斯编码参数推导
CN115209157B (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN114979656B (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN118803237A (zh) 视频编解码方法、装置、计算机可读介质及电子设备
WO2025118926A1 (zh) 视频编解码方法、装置、介质及电子设备
WO2025118925A1 (zh) 视频编解码方法、装置、介质及电子设备
CN120658875A (zh) 颜色分量的预测方法、装置、介质及电子设备
WO2025195257A1 (zh) 视频处理方法、装置、介质及电子设备
CN115209146B (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN114615497B (zh) 视频解码方法、装置、计算机可读介质及电子设备
CN115086656B (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN111770338B (zh) 编码单元的索引值确定方法、装置、设备及存储介质
WO2026026351A1 (zh) 视频编解码方法、装置及计算机可读介质
WO2025252185A1 (zh) 视频编解码方法、装置、计算机可读介质及电子设备
WO2025241790A1 (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN121967700A (zh) 视频编解码方法、装置、计算机可读介质及电子设备
CN121099052A (zh) 跨分量预测方法、视频编解码方法、装置、介质及设备
CN117834894A (zh) 视频编解码方法、装置、介质及电子设备
HK40092646A (zh) 视频编解码方法、装置、计算机可读介质及电子设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 25769212

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 1020267019671

Country of ref document: KR

Free format text: ST27 STATUS EVENT CODE: A-0-1-A10-A15-NAP-PA0105 (AS PROVIDED BY THE NATIONAL OFFICE)

WWE Wipo information: entry into national phase

Ref document number: 1020267019671

Country of ref document: KR

WWP Wipo information: published in national office

Ref document number: 1020267019671

Country of ref document: KR