WO2020182168A1

WO2020182168A1 - Decoding method and device

Info

Publication number: WO2020182168A1
Application number: PCT/CN2020/078874
Authority: WO
Inventors: 陈方栋
Original assignee: 杭州海康威视数字技术股份有限公司
Priority date: 2019-03-11
Filing date: 2020-03-11
Publication date: 2020-09-17
Also published as: CN110662031A; CN111698501A; WO2020182169A1; CN111698501B; CN110662031B

Abstract

The present application provides a decoding method and device. The decoding method comprises: determining, on the basis of decoded information, whether the current image block allows to enable a cross-component linear model prediction (CCLM) mode; if yes, determining a target prediction mode used by the current image block; if the target prediction mode is a CCLM reference mode, obtaining a reference pixel value according to the target prediction mode; determining a linear transformation parameter according to the reference pixel value; and determining a pixel value of the current component of the current image block according to the linear transformation parameter and a pixel value of another decoded component of the current image block.

Description

Decoding method and device

Technical field

This application relates to video coding and decoding technology, and in particular to a decoding method and device.

Background technique

The intra-frame prediction technology in video coding reduces the redundancy of the information to be coded and improves the compression efficiency of image data by using the correlation between pixels within a frame of image.

In H.265 intra-frame prediction, the luminance and chrominance share the prediction partition mode (part_mode) for independent prediction, that is, the luminance component uses the surrounding reconstructed luminance value to predict the luminance value of the current block to be coded, and the chrominance The prediction of the value is only based on the reconstructed chrominance values around it for intra prediction. However, studies have found that although the format of YUV (a color coding method) has reduced the correlation of different color components relative to RGB (a color mode), the chroma and brightness are still consistent in texture. The cross-component prediction technology uses the coded color component information to predict the color component to be coded, which further removes the correlation between the color components and improves the coding compression quality.

However, it has been found in practice that the current cross-component prediction technology does not limit the starting conditions of the cross-component prediction technology, which may lead to the use of the cross-component prediction technology in scenes that are not suitable for the use of the cross-component prediction technology, and thus lead to poor coding performance.

Summary of the invention

In view of this, the present application provides a decoding method and device.

According to a first aspect of the embodiments of the present application, a decoding method is provided, which includes: determining whether the current image block allows the cross-component linear model prediction (CCLM) mode to be enabled based on the decoded information; if the current image block allows the CCLM to be enabled Mode, determine the target prediction mode used by the current image block; if the target prediction mode is the CCLM reference mode, obtain the reference pixel value according to the target prediction mode; determine the linear transformation parameter according to the reference pixel value ; Determine the pixel value of the current component of the current image block according to the linear transformation parameter and the pixel value of another component of the current image block that has been decoded.

According to a second aspect of the embodiments of the present application, there is provided a decoding device, including a processor and a machine-readable storage medium, the machine-readable storage medium stores machine-readable instructions that can be executed by the processor, and The processor is prompted by the machine-readable instructions to: determine whether the current image block allows the CCLM mode to be enabled based on the decoded information; if the current image block allows the CCLM mode to be enabled, determine the target prediction mode used by the current image block; If the target prediction mode is the CCLM reference mode, obtain reference pixel values according to the target prediction mode; determine linear transformation parameters according to the reference pixel values; determine linear transformation parameters according to the linear transformation parameters and the current image block has been decoded The pixel value of the other component of, determines the pixel value of the current component of the current image block.

The decoding method of the embodiment of the present application determines whether the current image block allows the CCLM mode to be enabled based on the decoded information. When the current image block allows the CCLM mode to be enabled, the target prediction mode used by the current image block is determined, and when the target prediction mode is In CCLM reference mode, the reference pixel value is obtained according to the target prediction mode, and then the linear transformation parameter is determined according to the reference pixel value, and the current image block is determined according to the linear transformation parameter and the pixel value of another component of the current image block that has been decoded The pixel value of the current component of the CCLM mode is restricted by limiting the enabling conditions of the CCLM mode, which avoids the use of the CCLM mode in scenes where the CCLM mode is not applicable, and improves the coding performance.

Description of the drawings

Fig. 1 is a schematic diagram of a reference pixel value acquisition position in a CCLM_T mode;

Fig. 2 is a schematic diagram of a reference pixel value acquisition position in a CCLM_L mode;

Fig. 3 is a schematic flowchart of a decoding method shown in an exemplary embodiment of the present application;

Fig. 4A is a schematic diagram showing a reference pixel value acquisition position of a CCLM_L mode according to an exemplary embodiment of the present application;

Fig. 4B is a schematic diagram showing a reference pixel value acquisition position of a CCLM_T mode according to an exemplary embodiment of the present application;

Fig. 4C is a schematic diagram showing a reference pixel value acquisition position of a CCLM_T_L mode according to an exemplary embodiment of the present application;

Fig. 4D is a schematic diagram of obtaining linear transformation parameters according to an exemplary embodiment of the present application;

Fig. 5 is a schematic diagram of the hardware structure of a decoding device shown in an exemplary embodiment of the present application;

Fig. 6 is a schematic structural diagram of a decoding device shown in an exemplary embodiment of the present application.

detailed description

Here, exemplary embodiments will be described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same signs in different drawings indicate the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are only examples of devices and methods consistent with some aspects of the application as detailed in the appended claims.

The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit the application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

In order to make the foregoing objectives, features, and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

Please refer to FIG. 3, which is a schematic flowchart of a decoding method provided by an embodiment of this application, where the decoding method may be applied to a decoding end device. As shown in FIG. 3, the decoding method may include the following steps:

Step S300, based on the decoded information, determine whether the current image block allows the CCLM mode to be enabled. If yes, go to step S310; otherwise, end the current process.

Cross-Component Linear Model Prediction (CCLM for short) is a cross-component prediction technology, based on the area of the reference luminance block, it can include three CCLM reference modes: using the reconstructed value above the current chrominance block and the corresponding luminance area The schematic diagram of the CCLM_T mode of the current chrominance block can be shown in Figure 1; the reconstruction value on the left side of the current chrominance block and the CCLM_L mode corresponding to the luminance area are used, and the schematic diagram can be shown in Figure 2; The reconstruction value of and the CCLM_T_L mode of the corresponding brightness area.

In the embodiments of the present application, in order to improve the coding performance, the starting conditions of the CCLM mode may be limited to ensure that the scenes using the CCLM mode are all scenes suitable for using the CCLM.

As an example, when the decoding end device decodes the received code stream, it can determine whether the current image block allows the CCLM mode to be started based on the decoded information.

In a possible implementation manner of this application, when the current image block meets all the following conditions, it is determined that the current image block allows the CCLM mode to be enabled:

The prediction mode of the current image block is intra mode;

At least one other component of the current image block has decoded the required pixel information, and the current component exists and has not been decoded;

The prediction mode of the current image block is not the corresponding brightness component mode (Direct Mode, DM for short);

The decoding information of at least one side of the image block on the left side and the upper side of the current component of the current image block is available;

The size of the current image block is within the preset size range.

It should be noted that in the embodiment of the present application, the current component refers to the component being decoded in the current image block, which may include, but is not limited to, a luminance component or a chrominance component.

As an example, considering that the CCLM mode needs to be used in the intra mode, and the CCLM mode needs to refer to the decoding information of other components, and the decoding information of the image blocks on the left and upper sides of the current component, the current image can be Whether the block is in intra mode, whether there is at least one other component that has decoded the required pixel information, and whether the decoding information of at least one of the left and upper sides of the current component of the current image block is available as the current image block Reference conditions for whether to enable CCLM mode.

As an example, considering that the CCLM mode and the DM mode are mutually exclusive modes, that is, if the current image block uses the DM mode, the current image block does not allow the CCLM mode to be enabled. Therefore, the decoder device determines that the current image block is allowed to be enabled The CCLM mode needs to ensure that the prediction mode of the current image block is not the DM mode.

As an example, in order to ensure the performance of the CCLM mode and improve the coding performance, the size range of the image blocks that allow the CCLM mode to be enabled can also be limited (referred to herein as the preset size range, which can be set according to the actual scene).

As an example, the decoder device may determine that the prediction mode of the current image block is intra mode, at least one other component of the current image block has decoded the required pixel information, and the current component exists and is not decoded, and the prediction of the current image block When the mode is not DM mode, the decoding information of at least one of the image blocks on the left and upper side of the current component of the current image block is available, and the size of the current image block is within the preset size range, it is determined that the current image block is allowed to be enabled CCLM mode.

Exemplarily, in the embodiment of the present application, if the current component is a chrominance component, at least one other component may be a luminance component; if the current component is a luminance component, at least one other component may be a chrominance component; Is a Cb component, then at least one other component may be a Cr component.

It should be noted that, in the embodiments of this application, based on the reference conditions for determining whether the current image block allows the CCLM mode to be enabled, based on the reference conditions described in the foregoing implementation manners, the encoding end device may also determine the current image block based on the corresponding reference conditions. Whether the CCLM mode is allowed to be enabled when the component is encoded, its specific implementation is not repeated here.

As an example of this implementation manner, the size of the foregoing current image block is within a preset size range, which may include:

The area of the current image block is greater than or equal to the first area threshold and less than or equal to the second area threshold;

Wherein, the first area threshold and the second area threshold are both the n-th power of 2, n is a positive integer, and the second area threshold is greater than the first area threshold.

As an example, when determining whether the current image block allows the CCLM mode to be enabled, the size of the current image block can be characterized based on the area of the current image block, that is, it is necessary to ensure that the area of the current image block is within the preset area range. To allow the current image block to enable CCLM mode.

Exemplarily, the preset area range may be greater than or equal to the first area threshold (may be called Smin) and less than or equal to the second area threshold (Smax), that is, it is necessary to ensure that the area of the current image block is in the range of [Smin, Smax] Within; Smin≤Smax; Smin and Smax are the nth power of 2, and n is a positive integer.

As another example of this implementation, the size of the foregoing current image block is within a preset size range, which may include:

The width of the current image block is greater than or equal to the first width threshold and less than or equal to the second width threshold;

The height of the current image block is greater than or equal to the first height threshold and less than or equal to the second height threshold;

Among them, the first width threshold, the second width threshold, the first height threshold, and the second height threshold are the n-th power of 2, where n is a positive integer, the first width threshold is less than the second width threshold, and the first height threshold is less than the second Height threshold.

As an example, when determining whether the current image block allows the CCLM mode to be enabled, the size of the current image block can be characterized based on the width and height of the current image block, that is, it is necessary to ensure that the width and height of the current image block are at the preset width respectively. Within the range and the preset height range, the current image block is allowed to enable CCLM mode.

Exemplarily, the preset width range may be greater than or equal to the first width threshold (may be called Wmin), and less than or equal to the second width threshold (Wmax), that is, it is necessary to ensure that the width of the current image block is in the range of [Wmin, Wmax] Within; Wmin≤Wmax; Wmin and Wmax are the nth power of 2, and n is a positive integer.

Exemplarily, the preset height range may be greater than or equal to the first height threshold (may be called Hmin) and less than or equal to the second height threshold (Hmax), that is, it is necessary to ensure that the height of the current image block is in the range of [Hmin, Hmax] Within; Hmin≤Hmax; Hmin and Hmax are the nth power of 2, and n is a positive integer.

It should be realized that, in the embodiment of the present application, the above-mentioned characterizing the size of the image block by the area of the image block or characterizing the size of the image block by the width and height of the image block is only the representation of the image block in the embodiment of the present application. The specific example of the size is not a limitation of the protection scope of this application. For example, in the embodiment of the present application, the size of the image block may also be characterized by the area, width, and height of the image block, and its specific implementation is not described here.

In the embodiment of the present application, if the decoding end device determines that the current image block allows the CCLM mode to be enabled, the decoding end device may continue to decode the current component of the current image block in the manner of steps S310 to S340.

If the decoding end device determines that the current image block does not allow the CCLM mode to be enabled, the decoding end device can continue to decode the current image block according to other strategies. The specific implementation can refer to the related implementation of the related solution, which is not done here in the embodiment of the application. Repeat.

Step S310: Determine the target prediction mode used by the current image block.

In the embodiment of this application, when the decoder device determines that the current image block allows the CCLM mode to be enabled, the decoder device can analyze the specific prediction mode used by the current image block (referred to herein as the target prediction mode) to determine the current image block used Whether the target prediction mode is the CCLM reference mode (CCLM_L, CCLM_T, or CCLM_T_L).

In a possible implementation manner of the present application, the foregoing determination of the target prediction mode used by the current image block may include:

Determine a list of candidate prediction modes for the current image block;

Analyze the index value of the prediction mode used by the current image block;

Based on the parsed index value, the target prediction mode used by the current image block is determined in the candidate prediction mode list.

As an example, in order to determine the target prediction mode used by the current image block, on the one hand, the decoder device can determine the candidate prediction mode list of the current image block, that is, the prediction mode supported by the current image block; on the other hand, the decoder device can The index value of the prediction mode used by the current image block is parsed from the code stream, and then, based on the parsed index value, the target prediction mode used by the current image block is determined in the candidate prediction mode list.

As an example of this implementation manner, the foregoing determination of the candidate prediction mode list of the current image block may include:

Determine the candidate prediction mode list of the current image block as the default list.

As an example, a default list as a candidate prediction mode list can be pre-configured on the encoding end device and the decoding end device. The default list can be negotiated and determined by the encoding end device and the decoding end device, or manually configured by the user on the encoding end device. And decoding end equipment.

In an example, the above-mentioned default list is the first list or the second list including part of the prediction modes in the first list. Exemplarily, the first list is {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}. Exemplarily, the non-CCLM mode may include, but is not limited to, angular prediction mode, DM mode, Planar mode, or DC mode.

It should be noted that the order of the modes in the first list may be the order of the modes listed in the list, or other orders. This article takes the order listed in the list as an example.

For example, the second list may be {non-CCLM mode, CCLM_L mode, CCLM_T mode}, {CCLM_T_L mode, non-CCLM mode, CCLM_T mode} or {CCLM_T_L mode, CCLM_T mode}, etc.

As another example of this implementation manner, the foregoing determination of the candidate prediction mode list of the current image block may include:

Based on the location information of the current image block, a list of candidate prediction modes for the current image block is determined.

As an example, the decoding end device may determine the candidate prediction mode list of the current image block based on the location information of the current image block.

Exemplarily, the position information of the current image block may include: whether the decoding information of the left or upper image block of the current image block is available and the position of the current image block in the largest coding unit (LCU).

For example, assuming that the initial candidate prediction mode list of the current image block is {non-CCLM mode, CCLM_L mode, CCLM_T mode}, the decoder device can determine whether the decoding information of the left or upper image block of the current image block is available, and The position of the current image block in the LCU.

If the left information of the current image block is not available, the CCLM_T_L mode and CCLM_L mode are excluded from the candidate prediction mode list; if the upper information of the current image block is not available, the CCLM_T_L mode and CCLM_T mode are removed from the candidate prediction mode list Exclude: If the current image block is at the upper boundary of the LCU, that is, the current image block is connected to the upper LCU, the CCLM_T_L mode and the CCLM_T mode are excluded.

Exemplarily, the unavailability of the left or upper image block of the current image block includes, but is not limited to, the left or upper image block of the current image block does not exist, or exists but is not decoded.

As an example, when the decoding end device determines the candidate prediction mode list of the current image block, it may analyze the index value of the prediction mode used by the current image block.

Exemplarily, assuming that the total number of prediction modes in the candidate prediction mode list of the current block is Max, the decoding end device can parse Max-1 bits to determine the index value of the prediction mode used by the current image block.

For example, suppose Max=4 and the end bit is 1 (that is, when the decoding end device analyzes the binary codeword sequence corresponding to the index value of the prediction mode used by the current image block, when it resolves to 1, the code stream analysis ends) , The decoding device can parse up to 3 bits, or stop parsing immediately when the parsing reaches 1. At this time, if it is resolved to "1", the index value is determined to be 0; if it is resolved to "01", the index value is determined to be 1; if it is resolved to "001", the index value is determined to be 2; if it is resolved to "000" ", the index value is determined to be 3.

As an example, when the decoding end device parses the index value of the prediction mode used by the current image block, it may determine the target prediction mode used by the current image block in the determined candidate prediction mode list based on the index value.

It should be noted that, in the embodiment of the present application, if the decoding end device determines that the target prediction mode used by the current image block is the CCLM reference mode (CCLM_L, CCLM_T, or CCLM_T_L), the decoding end device can continue to follow the steps S320 to S340 Perform the decoding of the current component of the current image block.

If the decoding-end device determines that the target prediction mode used by the current image block is a non-CCLM reference mode, the decoding-end device can continue to decode the current image block according to other strategies, and its specific implementation can refer to related implementations of related solutions. This application implements The examples are not repeated here.

Step S320: If the target prediction mode is the CCLM reference mode, obtain the reference pixel value according to the target prediction mode.

In the embodiment of the present application, when the decoder device determines that the target prediction mode used by the current image block is the CCLM reference mode (CCLM_L, CCLM_T, or CCLM_T_L), the decoder device can determine the current component and the current image block according to the target prediction mode. The other component of the decoding obtains the reference pixel value.

In a possible implementation manner of the present application, the foregoing obtaining the reference pixel value according to the target prediction mode may include:

If the size of the current component of the current image block is the same as that of another component that has been decoded, the same position and the same number of pixel values are obtained;

If the size of the current component of the current image block is different from that of another component that has been decoded, sampling is performed according to the aspect ratio to obtain the same number of pixel values.

For example, suppose that the width of the current image block is width and the height is height.

If the width of the current component is 1/W of another component that has been decoded, the height of the current component is 1/H of the other component that has been decoded.

If the target prediction mode is the CCLM_L mode, the decoder device can take the left L reference pixel values from the current component, and the reference decoded other component takes the L reference pixels at the corresponding position after H times upsampling. value.

Exemplarily, L is 2*height.

If the target prediction mode is the CCLM_T mode, the decoder device can take the upper T reference pixel values from the current component, and the reference decoded other component takes the corresponding position and the corresponding position after W times upsampling. value. Exemplarily, T is 2*width.

If the target prediction mode is the CCLM_T_L mode, the decoding device can take the L reference pixel values on the left side and T reference pixel values on the upper side for the current component, and the decoded reference component takes the left side after H times upsampling. The L reference pixel values of the corresponding position and the upper side take the T reference pixel values of the corresponding position after W times upsampling. Exemplarily, L is height and T is width.

Step S330: Determine linear transformation parameters according to the obtained reference pixel value.

In the embodiment of the present application, when the decoding end device obtains the reference pixel value, it may determine the linear transformation parameter according to the obtained reference pixel value.

In a possible implementation manner of the present application, determining the linear transformation parameter according to the obtained reference pixel value may include:

Determine the maximum value maxL and minimum value minL of the reference pixel value of another component that has been decoded, and the maximum value maxC and minimum value minC of the reference pixel value of the current component;

According to maxL, minL, maxC and minC, determine the linear transformation parameters.

As an example, in order to determine the linear transformation parameters, the decoding device may first determine the maximum value (referred to as maxL in this article) and minimum value (referred to in this article as minL) of the reference pixel value of another component that has been decoded, and the current The maximum value (referred to herein as maxC) and the minimum value (referred to herein as minC) of the reference pixel value of the component.

After the decoder device determines maxL, minL, maxC, and minC, it can determine the linear transformation parameters based on maxL, minL, maxC, and minC.

In an example, the foregoing determination of linear transformation parameters according to maxL, minL, maxC, and minC includes:

If maxL is equal to minL, then α=0, β=Si;

Among them, α and β are linear transformation parameters, and Si is the average value of maxC, minC or maxC and minC.

As an example, if maxL is equal to minL, that is, the reference pixel values of another component that have been decoded are all the same, the decoding end device can determine the linear transformation parameter α=0, β=Si. Exemplarily, Si=maxC; or, Si=minC; or, Si=(maxC+minC)/2.

In another example, the foregoing determination of linear transformation parameters according to maxL, minL, maxC, and minC includes:

If maxL is not equal to minL, determine α and β according to the following formula:

diffL=maxL-minL;

diffC=maxC-minC;

x=Floor(Log2(diffL));

normDiff=((diffL<<4)>>x)&15, if normDiff! =0, then x=x+1; otherwise, x remains unchanged;

y=Floor(Log2(Abs(diffC)))+1;

add=(1<<y)>>1, value=DivSigTable[normDiff]|8

α=(diffC*value+add)>>y;

shift=3+x–y;

If shift<1, set shift=1 and α=Sign(α)*15;

β=minC-((α*minL)>>shift)

Among them, Floor means rounding down, Log2 means taking the logarithm of 2, a<<x means a left shift by x bits, >> means right shift, & means bitwise AND, | means bitwise OR, abs means absolute value , DivSigTable is a table, DivSigTable[]={0,7,6,5,5,4,4,3,3,2,2,1,1,1,1,0}, Sign(x) means x The sign bit of (if x is equal to 0, then Sign(x) is 0; if x is greater than 0, then Sign(x) is 1; otherwise, Sign(x) is -1).

As an example, taking the current component as a chrominance component and another decoded component as a luminance component as an example, the linear transformation parameters α and β can be determined according to the following formula:

β=C _j -α*L _j

As shown in Figure 4D, by traversing the reference pixel value of the decoded luminance component of the current image block, the maximum and minimum values of the reference pixel value of the decoded luminance component are determined, and according to the reference pixel value of the decoded luminance component The maximum and minimum values of, and the reference pixel value of the corresponding chrominance component, determine the linear transformation parameters.

In the above formula, Li is the maximum value of the reference pixel value of the decoded luminance component, Lj is the minimum value of the reference pixel value of the decoded luminance component; Ci and Cj are the reference chrominance components corresponding to Li and Lj, respectively Pixel values.

Considering that if the linear transformation parameters α and β are determined according to the above formula, more multiplication and division operations are required, and the current hardware is less friendly to multiplication and division operations (especially division operations). Therefore, according to The above formula determines the hardware friendliness of linear transformation parameters will be relatively poor.

Therefore, in the embodiment of the present application, data can be multiplied and divided by shifting, so as to improve the hardware friendliness of the determination of linear transformation parameters.

As an example, when the decoder device determines maxL, minL, maxC, and minC, and determines that maxL is not equal to minL, the decoder device can determine the linear transformation parameters α and β according to the following formula:

diffL=maxL-minL;

diffC=maxC-minC;

x=Floor(Log2(diffL));

y=Floor(Log2(Abs(diffC)))+1;

add=(1<<y)>>1, value=DivSigTable[normDiff]|8

α=(diffC*value+add)>>y;

shift=3+x–y;

If shift<1, set shift=1 and α=Sign(α)*15;

β=minC-((α*minL)>>shift)

For example, suppose diffL=3 (binary is 11), then diffL<<4 is 110000 (decimal is 48); suppose diffL=15 (binary is 1111), then diffL>>1 is 111 (decimal is 7), diffL>>2 is 11 (decimal is 3).

DivSigTable[normDiff] represents taking the normDiff bit in DivSigTable[]={0,7,6,5,5,4,4,3,3,2,2,1,1,1,1,0}.

For example, assuming normDiff=1, DivSigTable[normDiff] is 7; assuming normDiff=2, DivSigTable[normDiff] is 6; assuming normDiff=4, DivSigTable[normDiff] is 5.

It can be seen that in this embodiment, the multiplication and division operations are implemented by shifting left and shifting right, which improves the hardware friendliness.

Step S340: Determine the pixel value of the current component of the current image block according to the determined linear transformation parameter and the pixel value of another component of the current image block that has been decoded.

In this embodiment of the application, when the decoder device determines the linear transformation parameter, it can determine the pixel value of the current component of the current image block according to the determined linear transformation parameter and the pixel value of another component of the current image block that has been decoded.

In a possible implementation manner of the present application, the foregoing determination of the pixel value of the current component of the current image block according to the determined linear transformation parameter and the pixel value of another component of the current image block that has been decoded can be implemented by the following formula :

Yi=(αXi)＞＞z+β

Among them, Xi is the ith pixel value in another component that has been decoded, Yi is the ith pixel value in the current component, and z is a natural number.

As an example, when the decoding end device determines the linear transformation parameters α and β, the decoding end device can obtain the i-th pixel value of another component decoded in the current image block (referred to as Xi in this article), and according to α, β and Xi, the above formula Yi=(αXi)>>z+β is used to determine the i-th pixel value of the current component of the current image block (herein referred to as Yi).

Exemplarily, z=shift (the value of shift can be determined in the manner described in the above embodiment); or, z=0.

It should be noted that in the embodiments of the present application, the pixel value of another component of the current image block that has been decoded may be referred to as the first reference pixel value, and the pixel value of the current component of the current image block may be referred to as the target pixel value. The reference pixel value on the left or upper side of the current component of the image block may be called the second reference pixel value, and the reference pixel value on the left or upper side of the decoded other component of the current image block may be called the third reference pixel value.

Exemplarily, the aforementioned first reference pixel value, second reference pixel value, and third reference pixel value may be reconstruction values; the target pixel value may be a predicted value.

Or, exemplarily, the first reference pixel value, the second reference pixel value, and the third reference pixel value may be residual values; the target pixel value is a predicted value.

As an example, considering that in actual application scenarios, the pixel value of a certain component of an image block usually has a pixel value range (such as 0 to 255). Therefore, after the decoder device determines the pixel value of the current component in the above manner , It is also necessary to determine whether the pixel value of the current component is within the preset pixel value range.

As an example of this implementation manner, after the above determining the pixel value of the current component of the current image block, it may further include:

If the determined pixel value of the current component is greater than the first preset pixel value threshold, the first preset pixel value threshold is used as the pixel value of the current component of the current image block;

If the determined pixel value of the current component is less than the second preset pixel value threshold, the second preset pixel value threshold is used as the pixel value of the current component of the current image block;

Exemplarily, the first preset pixel value threshold is greater than the second preset pixel value threshold.

As an example, for any pixel of the current image block, after the decoder device determines the pixel value of the pixel in the above manner, if the pixel value is greater than the first preset pixel value threshold (which can be set according to requirements, such as 2 to the power of D-1, D represents the number of bits of the current pixel value. For example, when D is 8, the first preset pixel value threshold is 255), then the decoder device can determine the pixel value of the pixel as the first The preset pixel value threshold.

If the pixel value is less than the second preset pixel value threshold (such as 0, or it can be set to other values greater than 0 according to requirements), the decoder device can determine the pixel value of the pixel as the second preset pixel Threshold.

For example, assuming that the first preset pixel value threshold is 255 and the second preset pixel value threshold is 0, when the decoder device determines that the pixel value of the current component is 300, the decoder device can use 255 as the current component Pixel value; when the decoder device determines that the pixel value of the current component is -2, the decoder device can use 0 as the pixel value of the current component.

For another example, assuming that the first preset pixel value threshold is 255, and the second preset pixel value threshold is 2, then when the decoder device determines that the pixel value of the current component is 300, the decoder device may use 255 as the current component When the decoder device determines that the pixel value of the current component is 1, the decoder device can use 2 as the pixel value of the current component.

It should be noted that, in the embodiment of this application, for any CCLM reference mode (CCLM_L, CCLM_T, or CCLM_T_L), when the encoding end device determines the rate-distortion cost of the CCLM reference mode or uses the CCLM reference mode for the current image block When encoding the current component of the image block, the pixel value of the current component of the image block can be determined in the manner described in the above step S320 to step S340, and the specific implementation is not repeated here.

In order to enable those skilled in the art to better understand the technical solutions provided by the embodiments of the present application, the technical solutions provided by the embodiments of the present application will be described below with reference to specific examples.

As an example, when the current component of the current image block needs to be decoded, it can be implemented according to the following steps:

1. The decoding device can obtain the decoded information and determine whether the current image block meets the following conditions:

1.1. The prediction mode of the current image block is intra mode;

1.2. At least one other component of the current image block has decoded the required pixel information, and the current component exists and has not been decoded;

1.3. The prediction mode of the current image block is not DM mode;

1.4. The decoding information of at least one of the image blocks on the left side and the upper side of the current component of the current image block is available;

1.5. The size of the current image block is within the preset size range.

In this embodiment, for step 1.5, the following implementation manners may be included:

Example one

As an example, the size of the current image block is within the preset size range, which may include:

The area of the current image block is within the range of [Smin, Smax], Smin and Smax are the nth power of 2, and n is a positive integer.

Exemplarily, Smin=8 and Smax=4096.

Example two

The width of the current image block is within the range of [Wmin, Wmax];

The height of the current image block is within the range of [Hmin, Hmax];

Wmin, Wmax, Hmin, and Hmax are 2 to the nth power, and n is a positive integer.

Exemplarily, Wmin=2, Wmax=64, Hmin=2, Hmax=64.

Example three

The width of the current image block is within the range of [Wmin, Wmax];

The height of the current image block is within the range of [Hmin, Hmax];

The area of the current image block is within the range of [Smin, Smax];

Wmin, Wmax, Hmin, Hmax, Smin, and Smax are 2 to the nth power, and n is a positive integer.

Exemplarily, Wmin=2, Wmax=64, Hmin=2, Hmax=64, Smin=8, Smax=4096.

It can be seen that, in this embodiment, by limiting the enabling conditions of the CCLM mode, the use of the CCLM mode in scenarios where the CCLM mode is not applicable is avoided, and the coding performance is improved.

In the following, the decoding end device determines that the current image block allows the CCLM mode to be enabled as an example.

2. Determine the target prediction mode used by the current image block.

2.1. Determine the candidate prediction mode list of the current image block.

1), the list of candidate prediction modes is the default list

Example four

As an example, the default list may be (in order) {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}.

Example five

As an example, the default list may be part of the modes in {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}.

For example, the default list may be (in order) {CCLM_T_L mode, non-CCLM mode, CCLM_T mode}.

For another example, the default list may be (in order) {non-CCLM mode, CCLM_T mode}.

2) Determine the candidate prediction mode list of the current image block based on the position information of the current image block.

Exemplarily, the position information of the current image block may include: whether the decoding information of the left or upper image block of the current image block is available and the position of the current image block in the LCU.

2.2. Analyze the index value of the prediction mode used by the current image block.

For example, suppose the list of candidate prediction modes is (in order) {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}, and the end bit is 1, then the decoder device can parse up to 3 bits, or parse to 1. Then stop parsing immediately. At this time, if it is resolved to "1", the index value is determined to be 0; if it is resolved to "01", the index value is determined to be 1; if it is resolved to "001", the index value is determined to be 2; if it is resolved to "000" ", the index value is determined to be 3.

For another example, suppose that the list of candidate prediction modes is (in order) {CCLM_T_L mode, non-CCLM mode, CCLM_T mode}, and the end bit is 1, then the decoding end device can parse up to 2 bits, or immediately if it parses to 1 Stop parsing. At this time, if it is resolved to "1", the index value is determined to be 0; if it is resolved to "01", the index value is determined to be 1; if it is resolved to "00", the index value is determined to be 2.

2.3. Determine the target prediction mode used by the current image block from the candidate prediction mode list according to the parsed index value.

For example, suppose that the candidate prediction mode list is (in order) {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}, if the resolved index value is 0, the target prediction mode is CCLM_T_L mode; If the index value is 2, the target prediction mode is CCLM_L mode.

For another example, suppose that the candidate prediction mode list is (in order) {CCLM_T_L mode, non-CCLM mode, CCLM_T mode}, if the parsed index value is 0, then the target prediction mode is CCLM_T_L mode; if the parsed index value Is 2, the target prediction mode is CCLM_T mode.

In the following, the target prediction mode is the CCLM reference mode as an example.

3. Obtain reference pixel values according to the target prediction mode.

As an example, when the decoding end device determines the target prediction mode used by the current image block, it can obtain the reference pixel value from the current component of the current image block and another decoded component based on the target prediction mode.

Assuming that the width of the current component is W and the height is H, the width of the current component is 1/2 of another component that has been decoded, and the height of the current component is 1/2 of another component that has been decoded.

Example Six

Assume that the target prediction mode is CCLM_L mode.

As an example, the decoding end device can take 2H reference pixel values from the left side of the current component, and take 2H reference pixel values from the corresponding position on the left side of another component that has been decoded after being upsampled twice. It can be as shown in Figure 4A.

Exemplarily, the reference pixel value on the left side of the current component (taking the current component as a chrominance component as an example) may be the pixel value of the first, second...2H chrominance block on the left side of the current image block; The reference pixel value of the corresponding position on the left side of the other component (taking the luminance component as an example) after 2 times of upsampling is the first, third, fifth...2nd of the left side of the other component* Pixel values of 2H-1 luminance blocks.

Example Seven

Assume that the target prediction mode is CCLM_T mode.

As an example, the decoding device can take 2W reference pixel values from the upper side of the current component, and take 2W reference pixel values from the corresponding position of the upper side of the decoded another component after 2 times upsampling. The schematic diagram It can be as shown in Figure 4B.

Exemplarily, the reference pixel value on the upper side of the current component (taking the current component as a chrominance component as an example) may be the pixel value of the first, second...2Wth chrominance block on the upper side of the current image block; The reference pixel value of the corresponding position on the upper side of one component (take the luminance component as an example) after 2 times upsampling is the first, third, fifth... 2nd*2W of the other component’s upper side -1 The pixel value of a luminance block.

Example eight

Assume that the target prediction mode is CCLM_T_L mode.

As an example, the decoding end device can take H reference pixel values from the left side of the current component and W reference pixel values from the upper side of the current component; the decoded left side of another component is up-sampled by 2 times. The corresponding position takes H reference pixel values, and the corresponding position after 2 times upsampling from the upper side of the decoded another component takes W reference pixel values. The schematic diagram may be as shown in FIG. 4C.

Exemplarily, the reference pixel value on the left side of the current component (taking the current component as the chrominance component as an example) is the pixel value of the first, second...Hth chrominance block on the left side of the current image block. The reference pixel value of can be the pixel value of the first, second...Wth chrominance block on the upper side of the current image block; the left side of the other component (taking the luminance component as an example) is upsampled twice The reference pixel value of the corresponding position is the pixel value of the 1st, 3rd, 5th...2*H-1 luminance block on the left side of the other component, and the corresponding upper side after 2 times upsampling The reference pixel value of the position is the pixel value of the first, third, fifth...2*W-1 luminance block above the other component.

4. Determine the linear transformation parameters α and β according to the obtained reference value.

As an example, the decoder device may first determine maxL, minL, maxC, and minC, and determine the linear transformation parameters according to maxL, minL, maxC, and minC.

Example 9

As an example, determining the linear transformation parameters according to maxL, minL, maxC, and minC may include:

If maxL is equal to minL, then α=0, β=Si;

Among them, α and β are linear transformation parameters.

Exemplarily, Si=maxC; or, Si=minC; or, Si=(maxC+minC)/2.

diffL=maxL-minL;

diffC=maxC-minC;

x=Floor(Log2(diffL));

y=Floor(Log2(Abs(diffC)))+1;

add=(1<<y)>>1, value=DivSigTable[normDiff]|8

α=(diffC*value+add)>>y;

shift=3+x–y;

If shift<1, set shift=1 and α=Sign(α)*15;

β=minC-((α*minL)>>shift)

Exemplarily, Floor means rounding down, Log2 means taking the logarithm of 2, a<<x means a left shift by x bits, >> means right shift, & means bitwise AND, | means bitwise OR, abs means Absolute value, DivSigTable is a table, DivSigTable[]={0,7,6,5,5,4,4,3,3,2,2,1,1,1,1,0}, Sign(x) means Take the sign bit of x.

5. Determine the pixel value of the current component of the current image block according to the determined linear transformation parameters α and β and the pixel value of another component of the current image block that has been decoded.

As an example, when the decoder device determines the linear transformation parameters α and β, the pixel value of the current component of the current image block can be determined according to the determined linear transformation parameter and the pixel value of another component of the current image block that has been decoded. .

Example ten

As an example, the decoder device may determine the pixel value of the current component of the current image block by using the following formula:

Yi=(αXi)＞＞shift+β

In this formula, Xi is the value of the i-th pixel in another component that has been decoded, and Yi is the value of the i-th pixel in the current component. The value of shift can be determined by the formula in step 4 above.

As an example, after determining Yi, if Yi<0, set Yi to 0; if Yi is greater than 2 to the power of D minus 1 (taking Yi using D-bit encoding as an example), set Yi to 2. Subtract 1 to the power of D.

Example 11

Yi=(αXi)+β

In this formula, Xi is the i-th pixel value in another component that has been decoded, and Yi is the i-th pixel value in the current component.

In the embodiment of the present application, it is determined whether the current image block allows the CCLM mode to be enabled based on the decoded information. When the current image block allows the CCLM mode to be enabled, the target prediction mode used by the current image block is determined. When the target prediction mode is the CCLM reference mode, the reference pixel value is obtained according to the target prediction mode. Furthermore, based on the reference pixel value, the linear transformation parameter is determined. Then, the pixel value of the current component of the current image block is determined according to the linear transformation parameter and the pixel value of another component of the current image block that has been decoded. By limiting the enabling conditions of the CCLM mode, it is avoided to use the CCLM mode in scenarios where the CCLM mode is not applicable, and the coding performance is improved.

The method provided by this application is described above. The following describes the device provided by this application:

Please refer to FIG. 5, which is a schematic diagram of the hardware structure of a decoding apparatus provided by an embodiment of this application. The decoding apparatus may include a processor 501 and a machine-readable storage medium 502 storing machine-executable instructions. The processor 501 and the machine-readable storage medium 502 may communicate via a system bus 503. Furthermore, by reading and executing the machine executable instructions corresponding to the decoding control logic in the machine-readable storage medium 502, the processor 501 can execute the decoding method described above.

The machine-readable storage medium 502 mentioned herein may be any electronic, magnetic, optical, or other physical storage device, and may contain or store information, such as executable instructions, data, and so on. For example, the machine-readable storage medium can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard drive), solid state hard drive, any type of storage disk (Such as CD, DVD, etc.), or similar storage media, or a combination of them.

As shown in FIG. 6, the above-mentioned decoding control logic may include a first determination unit 610, a second determination unit 620, an acquisition unit 630, a third determination unit 640, and a fourth determination unit 650 in terms of functions.

The first determining unit 610 is configured to determine whether the current image block allows the CCLM mode to be enabled based on the decoded information;

The second determining unit 620 is configured to determine the target prediction mode used by the current image block if the CCLM mode is allowed to be enabled for the current image block;

The obtaining unit 630 is configured to obtain a reference pixel value according to the target prediction mode if the target prediction mode is the CCLM reference mode;

The third determining unit 640 is configured to determine linear transformation parameters according to the reference pixel value;

The fourth determining unit 650 is configured to determine the pixel value of the current component of the current image block according to the linear transformation parameter and the pixel value of another component of the current image block that has been decoded.

As a possible implementation manner, the first determining unit 610 is specifically configured to determine that the current image block allows the CCLM mode to be enabled when the current image block meets the following conditions:

The prediction mode of the current image block is an intra mode;

At least one other component of the current image block has decoded required pixel information, and the current component exists and has not been decoded;

The prediction mode of the current image block is not the DM mode;

Decoding information of at least one of the image blocks on the left side and the upper side of the current component of the current image block is available;

The size of the current image block is within a preset size range.

As a possible implementation manner, the size of the current image block is within a preset size range, including:

Wherein, the first area threshold value and the second area threshold value are both n-th power of 2, n is a positive integer, and the second area threshold value is greater than the first area threshold value.

Wherein, the first width threshold, the second width threshold, the first height threshold, and the second height threshold are the nth power of 2, where n is a positive integer, and the first width threshold is less than the A second width threshold, and the first height threshold is smaller than the second height threshold.

As a possible implementation manner, the second determining unit 620 is specifically configured to: determine the candidate prediction mode list of the current image block; analyze the index value of the prediction mode used by the current image block; and based on the index value, The target prediction mode used by the current image block is determined in the candidate prediction mode list.

As a possible implementation manner, the second determining unit 620 is specifically configured to use the default list as the candidate prediction mode list of the current image block.

As a possible implementation manner, the default list is a first list or a second list including some prediction modes in the first list;

Wherein, the first list is {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}.

As a possible implementation manner, the second determining unit 620 is specifically configured to determine the candidate prediction mode list of the current image block based on the position information of the current image block.

As a possible implementation manner, the position information of the current image block includes: whether the decoding information of the left or upper image block of the current image block is available and the position of the current image block in the LCU.

As a possible implementation manner, the acquiring unit 630 is specifically configured to: if the current component of the current image block has the same size as the other decoded component, acquire the same position and the same number of pixel values; If the size of the current component of the current image block is different from that of the other decoded component, sampling is performed in the aspect ratio to obtain the same number of pixel values.

As a possible implementation manner, the third determining unit 640 is specifically configured to determine the maximum value maxL and the minimum value minL of the reference pixel value of the decoded another component, and the maximum value maxC of the reference pixel value of the current component. And the minimum value minC; according to maxL, minL, maxC and minC, determine the linear transformation parameters.

As a possible implementation manner, the fourth determining unit 650 is further configured to: if the determined pixel value of the current component is greater than a first preset pixel value threshold, use the first preset pixel value threshold as the The pixel value of the current component of the current image block; if the determined pixel value of the current component is less than a second preset pixel value threshold, the second preset pixel value threshold is taken as the current component of the current image block Pixel values;

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply one of these entities or operations. There is any such actual relationship or order between. Moreover, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes those that are not explicitly listed Other elements of, or also include elements inherent to this process, method, article or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article, or equipment that includes the element.

The above descriptions are only examples of the application, and are not intended to limit the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included in the scope of protection of this application.

Claims

A decoding method applied to a decoding terminal device, the method including:

Based on the decoded information, determine whether the current image block is allowed to enable the cross-component prediction linear model and the cross-component linear model prediction CCLM mode;

If the current image block allows the CCLM mode to be enabled, determining the target prediction mode used by the current image block;

If the target prediction mode is the CCLM reference mode, obtain reference pixel values according to the target prediction mode;

Determining linear transformation parameters according to the reference pixel value;

Determine the pixel value of the current component of the current image block according to the linear transformation parameter and the pixel value of another component of the current image block that has been decoded.
The method according to claim 1, wherein when the current image block satisfies the following conditions, it is determined that the current image block allows the CCLM mode to be enabled:

The prediction mode of the current image block is an intra mode;

At least one other component of the current image block has decoded required pixel information, and the current component exists and has not been decoded;

The prediction mode of the current image block is not the corresponding brightness component mode DM mode;

Decoding information of at least one of the image blocks on the left side and the upper side of the current component of the current image block is available;

The size of the current image block is within a preset size range.
The method according to claim 2, wherein the size of the current image block is within a preset size range, comprising:

The area of the current image block is greater than or equal to the first area threshold and less than or equal to the second area threshold;

Wherein, the first area threshold value and the second area threshold value are both n-th power of 2, n is a positive integer, and the second area threshold value is greater than the first area threshold value.
The method according to claim 2 or 3, wherein the size of the current image block is within a preset size range, comprising:

The width of the current image block is greater than or equal to the first width threshold and less than or equal to the second width threshold;

The height of the current image block is greater than or equal to the first height threshold and less than or equal to the second height threshold;

Wherein, the first width threshold, the second width threshold, the first height threshold, and the second height threshold are all 2 to the nth power, n is a positive integer, and the first width threshold is less than The second width threshold, the first height threshold is less than the second height threshold.
The method according to claim 1, wherein the determining the target prediction mode used by the current image block comprises:

Determining a candidate prediction mode list of the current image block;

Parse the index value of the prediction mode used by the current image block;

Based on the index value, the target prediction mode used by the current image block is determined in the candidate prediction mode list.
The method according to claim 5, wherein the determining the candidate prediction mode list of the current image block comprises:

The default list is used as the candidate prediction mode list of the current image block.
The method according to claim 6, wherein the default list is a first list or a second list including some prediction modes in the first list;

Wherein, the first list is {CCLM_T_L mode, non-CCLM mode, CCLM_L mode, CCLM_T mode}.
The method according to claim 5, wherein the determining the candidate prediction mode list of the current image block comprises:

Based on the location information of the current image block, a candidate prediction mode list of the current image block is determined.
The method according to claim 8, wherein the position information of the current image block comprises:

Whether the decoding information of the left or upper image block of the current image block is available and the position of the current image block in the largest coding unit LCU.
The method according to claim 1, wherein the obtaining a reference pixel value according to the target prediction mode comprises:

If the size of the current component of the current image block is the same as that of the other decoded component, acquiring the same position and the same number of pixel values;

If the size of the current component of the current image block is different from that of the other decoded component, sampling is performed according to the aspect ratio to obtain the same number of pixel values.
The method according to claim 1, wherein the determining a linear transformation parameter according to the reference pixel value comprises:

Determining the maximum value maxL and the minimum value minL of the reference pixel value of the decoded another component, and the maximum value maxC and the minimum value minC of the reference pixel value of the current component;

According to maxL, minL, maxC and minC, the linear transformation parameters are determined.
The method according to claim 1, after the determining the pixel value of the current component of the current image block, further comprising:

If the determined pixel value of the current component is greater than the first preset pixel value threshold, use the first preset pixel value threshold as the pixel value of the current component of the current image block;

If the determined pixel value of the current component is less than the second preset pixel value threshold, the second preset pixel value threshold is used as the pixel value of the current component of the current image block,

Wherein, the first preset pixel value threshold is greater than the second preset pixel value threshold.
A decoding device includes a processor and a machine-readable storage medium, the machine-readable storage medium stores machine-readable instructions that can be executed by the processor, and the processor is prompted by the machine-readable instructions:

Based on the decoded information, determine whether the current image block allows the cross-component linear model prediction CCLM mode to be enabled;

If the current image block allows the CCLM mode to be enabled, determining the target prediction mode used by the current image block;

If the target prediction mode is the CCLM reference mode, obtain reference pixel values according to the target prediction mode;

Determining linear transformation parameters according to the reference pixel value;

Determine the pixel value of the current component of the current image block according to the linear transformation parameter and the pixel value of another component of the current image block that has been decoded.