WO2021203381A1

WO2021203381A1 - Video encoding and decoding method and apparatus, and computer-readable storage medium

Info

Publication number: WO2021203381A1
Application number: PCT/CN2020/084047
Authority: WO
Inventors: 马思伟; 孟学苇; 王苫社; 郑萧桢
Original assignee: 北京大学; 深圳市大疆创新科技有限公司
Priority date: 2020-04-09
Filing date: 2020-04-09
Publication date: 2021-10-14
Also published as: CN112514382A

Abstract

A video encoding and decoding method and apparatus, and a computer-readable storage medium. The video encoding and decoding method comprises: filtering a chrominance component; and when the chrominance component and a luminance component corresponding thereto are respectively located at different sides of a virtual filtering boundary, prohibiting performing cross-component filtering on the filtered chrominance component by using the luminance component corresponding to the chrominance component.

Description

Video encoding and decoding method, device and computer readable storage medium

Technical field

The present disclosure relates to the field of image processing, and in particular, to a video encoding and decoding method, device, and computer-readable storage medium.

Background technique

In the field of image processing, the encoding end predicts the current image block, and sequentially transforms, quantizes and encodes the residuals between the current image block and the predicted image block into a bit stream, and sends it to the decoding end. The encoding end also performs inverse quantization, inverse transformation, and obtains the reconstruction residual. The reconstructed image block is obtained from the reconstruction residual and the predicted image block, and the reconstructed image block is loop-filtered. The filtered reconstructed image block is used as the reference image for inter-frame prediction. Piece. The decoding end decodes the code stream, performs inverse quantization and inverse transformation to obtain the residual, predicts the current image block to generate a predicted image block, obtains a reconstructed image block from the residual and the predicted image block, and performs loop filtering on the reconstructed image block , The reconstructed image block after filtering is used as a reference image block for inter-frame prediction, and is output to an external device at the decoding end.

There is room for improvement in the efficiency of existing video codecs.

Summary of the invention

One aspect of the present disclosure provides a video encoding and decoding method, including:

Filter the chrominance components;

When the chrominance component and its corresponding luminance component are located on different sides of the virtual filtering boundary, it is forbidden to use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.

Another aspect of the present disclosure also provides a video encoding and decoding device, including:

Memory, used to store executable instructions;

The processor is configured to execute the executable instructions stored in the memory to perform the following operations:

Filter the chrominance components;

In yet another aspect of the present disclosure, there is also provided a computer-readable storage medium, which stores executable instructions, which, when executed by one or more processors, can cause the one or more processing The device executes the above-mentioned video encoding and decoding method.

In another aspect of the present disclosure, a video encoding and decoding method is also provided, including:

Filter the chrominance components;

Use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component, wherein the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component. Shift value.

In yet another aspect of the present disclosure, a video encoding and decoding device is also provided, including:

Memory, used to store executable instructions;

Filter the chrominance components;

Use the luminance component corresponding to the chrominance component to perform cross-component filtering on the chrominance component after filtering, wherein the offset value of the virtual filtering boundary of the chrominance component is equal to the offset of the virtual filtering boundary of the luminance component value.

When the chroma component and its corresponding brightness component are on different sides of the virtual filter boundary, it is forbidden to use the brightness component corresponding to the chroma component to perform cross-component filtering on the filtered chroma component, or to use the brightness corresponding to the chroma component The component performs cross-component filtering on the filtered chroma component, where the offset value of the virtual filter boundary of the chroma component is equal to the offset value of the virtual filter boundary of the luminance component. The present disclosure can reduce the cross-component adaptive loop filtering process. Time, improve filtering efficiency and save cache space.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the drawings that need to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the application. For those of ordinary skill in the art, without creative labor, other drawings can be obtained from these drawings.

Fig. 1 is a schematic diagram of a video encoding process according to an embodiment of the disclosure.

Fig. 2 is a schematic diagram of a video decoding process according to an embodiment of the disclosure.

Figure 3 shows the offset value of the virtual filter boundary of the luminance component.

Figure 4 shows the offset value of the virtual filter boundary of the chrominance component when the image sampling format is 420.

Figure 5 shows the offset value of the virtual filter boundary of the chrominance component when the image sampling format is 444 and 422.

Figure 6 shows image blocks with image sampling formats of 444, 422, and 420.

Figure 7a shows the chrominance component filter of adaptive loop filtering; Figure 7b shows the luminance component filter of adaptive loop filtering.

Figure 8a shows the processing of the adaptive loop filter at the virtual filter boundary of the luminance component; Figure 8b shows the processing of the adaptive loop filter at the virtual filter boundary of the chrominance component.

FIG. 9 is a schematic diagram of cross-component adaptive loop filtering according to an embodiment of the disclosure.

Fig. 10a is the filter shape of the cross-component adaptive loop filtering; Fig. 10b is the position of the filter of the cross-component adaptive loop filtering in the image block.

Figure 11 shows when the image sampling format is 420, the sampling ratio of the luminance component and chrominance component in the column direction and the offset value ratio of the virtual filtering boundary of the luminance component and chrominance component in the column direction.

Figure 12 shows the sampling ratio of the luminance component and chrominance component in the column direction and the offset value ratio of the virtual filter boundary of the luminance component and chrominance component in the column direction when the image sampling format is 444 and 422.

FIG. 13 is a flowchart of a video encoding and decoding method according to an embodiment of the disclosure.

Figure 14 shows the positional relationship between the chrominance component and its corresponding luminance component under the virtual filter boundary in the column direction.

Figure 15 shows another positional relationship between the chrominance component and its corresponding luminance component under the virtual filter boundary in the column direction.

Figure 16 shows another positional relationship between the chrominance component and its corresponding luminance component under the virtual filter boundary in the column direction.

Figure 17 shows another positional relationship between the chrominance component and its corresponding luminance component under the virtual filter boundary in the column direction.

Figure 18 shows another positional relationship between the chrominance component and its corresponding luminance component under the virtual filter boundary in the column direction.

FIG. 19 shows the positional relationship between chrominance components and their corresponding luminance components in a video encoding and decoding method according to another embodiment of the present disclosure.

Figure 20 shows the positional relationship between the chrominance component and its corresponding luminance component under the virtual filter boundary in the row direction.

FIG. 21 is a schematic diagram of a video encoding and decoding device according to an embodiment of the disclosure.

FIG. 22 is a flowchart of a video encoding and decoding method according to another embodiment of the present disclosure.

FIG. 23 shows the positional relationship between the chrominance components and their corresponding luminance components under the virtual filter boundary in the column direction in the video encoding and decoding method of another embodiment of the present disclosure.

FIG. 24 shows the positional relationship between chrominance components and their corresponding luminance components under the virtual filter boundary in the row direction in the video coding and decoding method of another embodiment of the present disclosure.

FIG. 25 is a schematic diagram of a video encoding and decoding device according to another embodiment of the present disclosure.

Detailed ways

The technical solutions of the present disclosure will be clearly and completely described below in conjunction with the embodiments and the drawings in the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present disclosure.

The embodiments of this application are applicable to, but not limited to, the international video coding standards H.264/MPEG-AVC, H.265/MEPG-HEVC, the domestic audio and video coding standards AVS2, and the H.266/VVC international standards under development and AVS3 domestic Standards, and other video codec standards that will evolve in the future.

A video image generally includes an image frame of three components. The three components include: a luma (Luma) component Y and two chroma (Chroma) components Cb/Cr or Cg/Co. The video encoding process is shown in Figure 1. The original image frame of each component is divided into image blocks, and the current image block is used as the image block to be encoded. According to different prediction modes, the encoding end performs intra prediction (Intra Coding) or inter prediction (Inter Coding) on the current image block based on the reference image block to generate a predicted image block. Transform the residual between the current image block and the predicted image block (Transform) to generate transform coefficients, quantize the transform coefficients (Quantization) to generate quantized coefficients, and the quantized coefficients are entropy coded (Entropy Coding) to generate encoded codes Stream, the code stream is sent to the decoding end.

In the video encoding process, the encoding end also performs an inverse quantization operation on the quantized coefficients to obtain transform coefficients, and the inverse quantization operation is the inverse operation of the quantization operation. Perform an inverse transform operation on the transform coefficients obtained by the inverse quantization operation to obtain the reconstruction residual. A reconstructed image block (also called a reconstructed image block) is obtained from the reconstruction residual and the predicted image block. The reconstructed image block is used as a reference image block for intra-frame prediction to perform intra-frame prediction on other image blocks. The reconstructed image block also passes through a filter, and the filter performs loop filtering on the reconstructed image block to obtain a filtered reconstructed image block. The filtered reconstructed image block is used as a reference image block for inter-frame prediction to perform inter-frame prediction on other image frames. The prediction mode, intra-frame prediction information, motion compensation information, filter coefficients and other information in the video encoding process are also encoded into the code stream by entropy encoding and sent to the decoding end.

The video decoding process is shown in Figure 2. The video decoding process can basically be regarded as the inverse process of the video encoding process. The decoding end performs Entropy Decoding on the code stream to obtain information such as quantization coefficients, prediction modes, intra-frame prediction information, motion compensation information, and filter coefficients. The inverse quantization operation is performed on the quantized coefficient to obtain the transform coefficient, and the inverse quantization operation is the inverse operation of the quantization operation. Perform an inverse transform operation on the transform coefficients obtained by the inverse quantization operation to obtain the residual. According to the received prediction mode, the decoding end performs intra-frame prediction or inter-frame prediction on the current image block based on the reference image block to generate a predicted image block. The decoding end then obtains the reconstructed image block from the residual and the predicted image block. The reconstructed image block is used as a reference image block for intra-frame prediction to perform intra-frame prediction on other image blocks. The reconstructed image block also passes through a filter, and the filter performs loop filtering on the reconstructed image block to obtain a filtered reconstructed image block. The filtered reconstructed image block is used as a reference image block for inter-frame prediction to perform inter-frame prediction on other image frames. The reference image block can also be composed of the decoded image frame and output to the external device of the decoding end, and the external device can be, for example, a display of the decoding end.

In the above video encoding and video decoding process, loop filtering includes: deblocking filter (DF, Deblocking Filter), adaptive sample compensation filter (SAO, Sample Adaptive Offset) and adaptive loop filter (ALF, Adaptive Loop Filter) ), cross-component adaptive loop filtering (CCALF, Cross-Component ALF). Among them, deblocking filtering is mainly used to eliminate blocking effects between images caused by encoding by different processors; SAO, ALF, and CCALF are mainly used to compensate for the distortion of original pixels and reconstructed pixels due to encoding.

Since the image frame is divided into image blocks for prediction and transform quantization, the difference in encoding parameters between adjacent image blocks may cause blocking artifacts in the encoded reconstructed image. In order to remove the block effect of the reconstructed image block, the reconstructed image block is first subjected to deblocking filtering. The deblocking filter corrects the pixel value of the reconstructed image block, especially the pixel value near the boundary of the reconstructed image block, so as to achieve the purpose of eliminating the block effect.

In the process of deblocking filtering, when processing the boundary of an image block (such as a CTU), the pixel values on both sides of the boundary need to be used, so there is a dependency between two adjacent image blocks on both sides of the boundary. In order to realize the parallel operation of the image block loop filter, the above-mentioned dependency needs to be removed, and a virtual filter boundary is set in the image block. The virtual filter boundary has an offset value in the first direction relative to the boundary between the coded image blocks. The offset value can also be called offset, misalignment, misalignment value, distance, distance value, etc., refers to the virtual filter boundary The offset value compared to the boundary between coded image blocks. For ease of presentation, all are referred to as offset values below. The offset value represents the number of pixels between the virtual filter boundary and the boundary of the coded image block. The virtual filter boundary may also be referred to as a virtual boundary. In an example, the first direction refers to the column direction of the coded image block. In the existing standard, the position 4 lines above the CTU boundary of the luminance component is used as the virtual filter boundary, and the position 2 lines above the CTU boundary of the chrominance component is used as the virtual filter boundary, so that the pixels on both sides of a virtual filter boundary are independent of each other. , Can achieve row-level parallelism. As shown in FIG. 3, FIG. 3 is a schematic diagram of the virtual filtering boundary of the luminance component. The pixels above the virtual filter boundary are represented by open circles, and the pixels between the virtual filter boundary and the boundary between the coded image blocks are represented by gray circles. The offset value of the virtual filter boundary of the luminance component is 4 rows, that is, the distance between the virtual filter boundary of the luminance component and the boundary between the coded image block of the luminance component is 4 rows of pixels. For the chrominance component, the offset value of the virtual filtering boundary is 2, that is, the distance between the virtual filtering boundary of the chrominance component and the boundary between the coded image block of the chrominance component is 2 rows of pixels. 4 is a schematic diagram of the virtual filter boundary of the chrominance component of the 420 image sampling format, and FIG. 5 is a schematic diagram of the virtual filter boundary of the chrominance component of the 422 image sampling format. By setting the above virtual filter boundary, the pixels on both sides of the virtual filter boundary are independent of each other, thereby realizing the parallel operation of the image block at the row level. The foregoing is only an exemplary description, and the first direction may also be other directions, such as the row direction of the coded image block.

The existing standards support several sampling formats of YUV4: 2: 0 (420), YUV 4: 2: 2 (422), YUV 4: 4: 4 (444), and YUV 4: 0:0 (400), 4: The 0:0 format has only luminance components and no chrominance components. YUV 4:4:4, YUV 4:2:2, YUV 4:2:0 format is shown in 6, solid dots represent brightness points, hollow circles represent chromaticity points, and the sampling ratio of three components in YUV4:4:4 same. YUV4: 2:2 means that the Y component and the UV component are sampled at a ratio of 2:1. There are 10 Y pixels horizontally, only 5 UV pixels, 10 Y pixels vertically, and 10 UV pixels. YUV4: 2: 0 sampling means that there are 10 Y component pixels horizontally, 5 UV pixels, 10 Y pixels vertically, and 5 UV points. As shown in Figure 6, the solid dots represent the luminance component pixels, and the open circles represent the chrominance component pixels. The sampling ratio of the luminance component and the chrominance component in the 444 format is the same. In the 422 format, the luminance component and the chrominance component are sampled at a ratio of 2:1 in the row direction, and sampled at a ratio of 1:1 in the column direction. In the 422 format in Figure 6, there are 6 luminance component pixels in the row direction and only 3 chrominance component pixels, and 5 luminance component pixels and 5 chrominance component pixels in the column direction. The luminance component and chrominance component of the 420 format are sampled at a ratio of 2:1 in both the row direction and the column direction. In the 420 format in Figure 6, there are 6 luminance component pixels and 3 chrominance component pixels in the row direction, and 5 luminance component pixels and 2 chrominance component pixels in the column direction.

The reconstructed image block after deblocking filtering is subjected to adaptive sample compensation filtering. The adaptive sample compensation filter classifies the pixels of the reconstructed image block and adds the same compensation value to each type of pixel, so that the reconstructed image block is closer to the current image block, thereby suppressing the ringing effect.

The reconstructed image block after adaptive sample compensation filtering is subjected to adaptive loop filtering (ALF). ALF is a Wiener filter, which is used to minimize the mean square error between the current image block and the reconstructed image block. The surrounding pixels of the current pixel are multiplied by the corresponding filter coefficients and then summed to obtain the filter value of the current pixel. For the luminance component, as shown in Figure 7b, the filter is a 7×7 diamond, that is, the center pixel is filtered using 7 surrounding pixels. The pixel corresponding to C12 is the pixel to be filtered. Multiply the pixel at the position of C0-C12 by the corresponding filter coefficient, and then add the products to obtain the ALF filtering result of the pixel at the position of C12. Among them, the filter coefficients can also be referred to as filter weights. For the chrominance component, as shown in Figure 7a, the filter is a 5×5 diamond, that is, the center pixel is filtered using 5 surrounding pixels. The pixel corresponding to C6 is the pixel to be filtered, and the C0-C6 pixels are multiplied by the respective corresponding filter coefficients, and the products are added together to obtain the ALF filtering result of the C6 pixel. The points used in this process are all points in the reconstructed frame obtained before ALF.

Of course, the above is only an exemplary description. For the luminance component and the chrominance component, filters of other shapes other than 7×7 rhombus and 5×5 rhombus may also be used.

ALF can classify the pixels of the reconstructed image block, and different types of pixels use different filter coefficients. There are many ways to classify pixels. In a classification method, the luminance component can be classified according to the Laplacian direction, and the chrominance component is not classified. According to the Laplace direction, the pixels of the brightness component can be divided into 25 categories, which corresponds to 25 sets of filter coefficients. For pixels belonging to the same category, the filter coefficients corresponding to the category are used for filtering.

Specifically, each 4x4 block in the reconstructed image block is classified according to the Laplacian direction

Among them, C represents the category of the 4x4 block; D represents the direction classification result;

Indicates the fine classification result after the direction classification.

D is calculated as follows:

Among them, i,j represent the coordinate position of the 4x4 block in the entire image block; R(k,l) represents the pixel value located at the (k,l) position in the _{4x4 block; V k,l} represents the 4x4 block located at (i, j) The Laplacian gradient of the pixel of the coordinate in the column direction; H _k,l represents the Laplacian gradient of the pixel located at the (i,j) coordinate in the 4x4 block in the row direction; D1 _k,l represents the Laplacian gradient of the pixel located in the 4x4 block (i, j) coordinates of pixels in the 135 degree Laplacian gradient; D2 _{k, l} represents the 4x4 block at the (i, j) coordinates of the pixel at 45 degrees Laplacian gradient; g _v represents the 4x4 block in the Laplacian gradient Laplacian gradient in the column direction; g _h represents the Laplacian gradient of the 4x4 block in the row direction; g _d1 represents the Laplacian gradient of the 4x4 block in the 135 degree direction; g _d2 represents the 4x4 block in the 45 degree direction Laplace gradient.

in,

Represents the maximum value of the Laplacian gradient in the row and column directions;

Represents the minimum value of the Laplacian gradient in the row and column directions;

Represents the maximum value of the Laplacian gradient in the 45 and 135 directions;

Represents the minimum value of the Laplacian gradient in the 45 and 135 directions; R _{h, v} represents the ratio of the Laplacian gradient in the row and column directions; R _{d0, d1} represents the ratio of the Laplacian gradient in the 45 and 135 directions.

if

and

D is set to 0;

if

and

D is set to 1;

if

and

D is set to 2;

if

and

D is set to 3;

if

and

D is set to 4;

Among them, t1 and t2 are preset thresholds.

Is calculated as follows:

Then quantize A to get an integer between 0 and 4, as

Value.

According to the above method, the pixels of the brightness component can be divided into 25 categories, which corresponds to 25 sets of filter coefficients.

ALF can be divided into linear adaptive loop filtering and nonlinear adaptive loop filtering. Linear ALF can be expressed by the following formula:

Among them, I(x+i, y+j) is the pixel value before ALF filtering, that is, the input of the ALF filter; O(x, y) represents the ALF filtered at the position of the reconstructed image block (x, y) The pixel value is the output of the ALF filter; w(i,j) represents the ALF filter coefficient; (i,j) represents the relative position of the pixel in the filter from the pixel to be filtered.

In order to ensure that all ALF filter coefficients are integers, linear ALF also amplifies all ALF coefficients, that is, amplifies the original ALF filter coefficients by 128 times and then converts them to integers.

Linear ALF can also be expressed by the following formula:

Among them, L represents the length of the ALF filter; for a 7x7 diamond, L=7; for a 5x5 diamond, L=5.

Since the sum of all ALF filter coefficients is 128, that is, before the amplification, the sum of all ALF filter coefficients is 1, the above linear ALF formula can be equivalently transformed into the following form:

O(x,y)=I(x,y)+∑ _{(i,j)≠(0,0)} w(i,j).(I(x+i,y+j)-I(x,y )).

The difference between nonlinear ALF and linear ALF is that nonlinear ALF introduces a clip operation, each set of filter coefficients corresponds to a clip parameter, and each clip parameter corresponds to an index, and the index is also encoded into the code stream by entropy coding , And send to the decoder.

Non-linear ALF is expressed by the following formula:

Among them, K (d, b) = min (b, max (-b, d) represents the operation of the clip, k (i, j) represents the parameters of the clip, each ALF filter coefficient corresponds to a parameter of the clip, the parameter Choose from the following values:

For the luminance component, select one from {1024, 181, 32, 6};

For chrominance components, select one from {1024, 161, 25, 4};

Each parameter value may correspond to an index value. For example, the index value of clip parameter 1024 is 0, and the index value of clip parameter 181 is 1. The index value is also encoded into the code stream by entropy coding and sent to the decoding end.

For the row-level parallel processing of image blocks, ALF cannot filter across virtual filter boundaries. For the luminance component, as shown in FIG. 8a, for the pixel C12 to be filtered, when the surrounding pixels are on the other side of the virtual filtering boundary, the pixels on the other side and the pixels at their geometrically symmetric positions are not used to filter C12. For example, for the two drawings in the first row in Fig. 8a, the pixel at C0 is located on the other side of the virtual filter boundary, the upper and lower pixels at C0 are not used, and the two pixels at C2 are not used. The filter coefficient is updated to c2+c0, where c2 is the original filter coefficient of the pixel at C2, and c0 is the original filter coefficient of the pixel at C0. Similarly, for the two drawings in the second row of Figure 8a, if the pixel at position C0-C3 is on the other side of the virtual filter boundary, then the filter coefficients of the two pixels at position C5 are updated to c5+c1, The filter coefficients of the two pixels at the C6 position are updated to c6+c2+c0, and the filter coefficients of the two pixels at the C7 position are updated to c7+c3. For the two drawings in the third row of Fig. 8a, if the pixel at position C0-C8 is on the other side of the virtual filter boundary, the filter coefficients of the two pixels at position C10 are updated to c10+c4+c8, and the The filter coefficients of the two pixels at the C11 position are updated to c11+c5+c7+c3+c1, and the filter coefficients of the two pixels at the C12 position are updated to c12+2*c6+2*c2+2*c0. Among them, c0-c12 represent filter coefficients or filter weights.

The chrominance component is processed in a similar way to the luminance component. As shown in Figure 8b, for the pixel C6 to be filtered, when the surrounding pixels are on the other side of the virtual filtering boundary, the pixels on the other side and their geometrically symmetrical pixel pairs are not used. C6 performs filtering. For example, if the pixel at C0 is on the other side of the virtual filter boundary, the upper and lower pixels at C0 are not used, and the filter coefficients of the two pixels at C2 are updated to c2+c0, where c2 is the C2 position The original filter coefficient of the pixel, c0 is the original filter coefficient of the pixel at C0. Similarly, if the pixels at C0-C3 are on the other side of the virtual filter boundary, the filter coefficients of the two pixels at C5 are updated to c5+c1+c3, and the filter coefficients of the two pixels at C6 are updated to c6+2*c2+2*c0.

CCALF uses the luminance component to filter the chrominance component to improve the quality of the chrominance component. As shown in Figure 9, after ALF, CCALF uses the luminance component before ALF (that is, the luminance component after SAO) to filter the chrominance component Cr/Cb after ALF, and the filter value of CCALF is used as the chrominance component Cr/Cb. ALF offset, and add the ALF offset to the ALF filter value of the chrominance component Cr/Cb, and use the added result as the loop filter result of the chrominance component Cr/Cb.

As shown in Fig. 10a and Fig. 10b, the shape of the CCALF filter is a 3x4 rhombus, and the chrominance component pixel corresponding to 2 is the pixel to be filtered. Use the luminance component pixels corresponding to 2 and the surrounding luminance component pixels 0-7 to multiply their respective CCALF filter coefficients, and then add the products to obtain the CCALF filtering result of the chrominance component pixels corresponding to 2. Each frame of image can include multiple sets of CCALF filter coefficients, the chrominance components Cb and Cr have their own CCALF filter coefficients, the number of CCALF coefficient sets corresponds to an index, and the index used by each reconstructed image block is also encoded into the code stream by entropy coding and sent To the decoding end.

For CCALF, when the image sampling format is 420, the sampling ratio of the luminance component and the chrominance component in the column direction is 2:1, as shown in Figure 11, the virtual filter boundary of the luminance component and the chrominance component in the column direction The offset value ratio is 4:2, that is, the sampling ratio is the same as the offset value ratio. For pixels with chroma components located on the upper side of the virtual filter boundary, the corresponding CCALF luminance component pixels are also located on the upper side of the virtual filter boundary, that is, color The degree component and its corresponding luminance component are all located on the same side of the virtual filtering boundary. As shown in Fig. 11, the luminance component pixels of CCALF corresponding to the chrominance component pixels A, B, C, and D are A1, B1, C1, and D1, respectively. Among them, A and B are located on the upper side of the virtual filtering boundary of the chrominance component, and A1 and B1 corresponding to A and B are also located on the upper side of the virtual filtering boundary of the luminance component. In this way, when performing parallel operations at the image block line level (for example, CTU line level), when performing CCALF filtering on the previous line of image blocks (ie, one line of CTU on the upper side of the virtual filter boundary), the SAO of the luminance component has been completed, so only The CCALF of all the chrominance component pixels on the upper side of the virtual filtering boundary can be realized through the image block in the upper row.

When the sampling ratio of the luminance component and the chrominance component in the column direction is different from the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the column direction, for example, when the image sampling format is 422 and 444, as shown in Figure 12 As shown, the sampling ratio of the luminance component and the chrominance component in the column direction is 1:1, and the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the column direction is 4:2, so that A and B are located at The chrominance component is on the upper side of the virtual filtering boundary, while A1 and B1 corresponding to A and B are located on the lower side of the luminance component virtual filtering boundary, that is, the chrominance component and its corresponding luminance component are located on different sides of the virtual filtering boundary. In this way, when performing CCALF filtering on the chrominance component of the previous line of image blocks, since A1 and B1 corresponding to A and B are located under the virtual filtering boundary of the luminance component, it is only necessary to wait until the next line of image blocks (ie, the lower side of the virtual filtering boundary). CCALF can be performed on the two chrominance components A and B only after the luminance component of one row of CTU) has been SAO. This adds extra waiting time in CCALF, reduces the filtering efficiency of CCALF, and increases video codec The time of the process. At the same time, since it is necessary to wait for the brightness component of the next line of image blocks to complete the SAO, a buffer is needed to buffer the ALF filter values of A and B during the waiting process. After the brightness component of the next line of image blocks completes the SAO, the ALF filter values of A and B in the buffer are read out for CCALF, so more buffer space of the buffer is needed. Therefore, the above-mentioned CCALF processing method has the defects of low filtering efficiency and large buffer space occupation.

An embodiment of the present disclosure provides a video encoding and decoding method. The video encoding and decoding method can be applied to a video encoding end or a video decoding end, and can be applied to a device that has both video encoding and video decoding functions. The above-mentioned problems of low filtering efficiency and large buffer space occupation. As shown in Figure 13, the video encoding and decoding method includes:

Step S1301: filtering the chrominance components.

Step S1302: When the chrominance component and its corresponding brightness component are located on different sides of the virtual filtering boundary, it is forbidden to use the brightness component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.

Among them, the loop filtering, adaptive loop filtering, cross-component adaptive loop filtering, etc. mentioned in the embodiments of this application may also have other names, and this application is only an example and is not limited.

The filtering of the chrominance component in step S1301 includes performing one or more of DF, SAO, or ALF on the reconstructed image block of the chrominance component to obtain a filtering result after the chrominance component is filtered. For example, filtering the chrominance component includes: performing ALF on the reconstructed image block of the chrominance component to obtain the filtering result after ALF.

In step S1302, the chrominance components and their corresponding brightness components are located on different sides of the virtual filtering boundary, and it is forbidden to use the brightness components corresponding to the chrominance components to perform cross-component filtering on the filtered chrominance components. Here, cross-component filtering can be Including cross-component adaptive loop filtering.

Exemplarily, the chrominance component and its corresponding luminance component are located on different sides of the virtual filter boundary, and may include the sampling ratio of the luminance component and the chrominance component in the first direction and the luminance component and the chrominance component in the first direction. When the ratio of the offset value of the virtual filter boundary in the direction is different.

Exemplarily, the correspondence described in this embodiment may also be referred to as association, reference, mapping, and so on. "Correspondence" can be understood as the luminance component that the chrominance component refers to when performing cross-component adaptive loop filtering, for example, the luminance component at the same position or the same group. For example, when the image sampling format is 444 or 422, the luminance component that the chrominance component refers to when performing cross-component adaptive loop filtering is the luminance component with the same position. For 420, the luminance component that the chrominance component refers to when performing cross-component adaptive loop filtering is the luminance component of the same group. For example, in FIG. 11, A1 is the brightness component of the same group as A, and B1 is the brightness component of the same group as B.

The offset value can also be called offset, misalignment, misalignment value, distance, distance value, etc. It refers to the offset value of the virtual filter boundary compared to the boundary between the coded image blocks, and represents the virtual filter boundary and the coded image block. The number of pixels between the borders. The coded image block may be a coding tree unit (CTU), a coding unit (CU) included in the CTU, a prediction unit (PU) and a transform unit (TU) included in the CU. For example, in the image column direction, for the luminance component, the offset value of the virtual filter boundary is 4, that is, 4 rows of pixels, and the virtual filter boundary of the luminance component is shifted by 4 rows of pixels compared to the CTU boundary.

In an example, the first direction refers to the column direction of the luminance component and the chrominance component. In other examples, the first direction may also refer to other directions of the luminance component and the chrominance component, such as the row direction.

The image sampling format includes 420, 422, 444, and any other possible sampling formats, and the sampling ratio can be 1:1, 2:1, and any other possible sampling ratios. The offset value of the virtual filter boundary of the luminance component in the column direction can include 4, and any other possible values, and the offset value of the virtual filter boundary of the chrominance component in the column direction can include 2, and any other possible values. , The offset value ratio can be 2:1, and any other possible values.

In this embodiment, when the sampling ratio of the luminance component and the chrominance component in the first direction is different from the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the first direction, it includes: the luminance component and the chrominance component in the first direction. The sampling ratio of the chrominance component in the first direction is smaller than the ratio of the offset value of the virtual filtering boundary of the luminance component and the chrominance component in the first direction. As long as the above sampling ratio is less than the offset value ratio, when the chrominance component and its corresponding brightness component are located on different sides of the virtual filtering boundary, this embodiment prohibits the use of the brightness component corresponding to the chrominance component to the filtered chrominance component. Proceed to CCALF.

In this embodiment, prohibiting the use of the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component includes: prohibiting the use of the luminance component that is located on a different side of the virtual filtering boundary from the chrominance component to the filtered chrominance component Perform cross-component filtering.

As an example, prohibiting the use of luminance components located on a different side of the virtual filtering boundary from the chrominance component to perform cross-component filtering on the filtered chrominance component includes:

CCALF is performed on the filtered chrominance component using the luminance component located on the same side of the virtual filtering boundary as the chrominance component.

In this embodiment, that the chrominance component and its corresponding brightness component are respectively located on different sides of the virtual filtering boundary means that the chrominance component and its corresponding brightness component are located on opposite sides of the virtual filtering boundary. For example, when the chrominance component is located on the upper side of the virtual filtering boundary, its corresponding luminance component is located on the lower side of the virtual filtering boundary; when the chrominance component is located on the lower side of the virtual filtering boundary, its corresponding luminance component is located on the virtual filtering boundary. On the upper side.

In the following, the image sampling format is 422 and 444, the sampling ratio in the column direction is 1:1, the offset value of the virtual filter boundary of the luminance component in the column direction is 4, and the offset value of the virtual filter boundary of the chrominance component in the column direction is 4. The shift value is 2 and the shift value ratio is 2:1 as an example, and this embodiment is further described.

Those skilled in the art can select the luminance component for the chrominance component according to the distance according to the actual situation. In an example, performing CCALF on the filtered chrominance component using the luminance component on the same side of the virtual filtering boundary as the chrominance component includes:

CCALF is performed on the filtered chrominance component using the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component and is closest to the chrominance component.

When there are multiple chrominance components that are located on different sides of the virtual filtering boundary by themselves and their corresponding luminance components, a luminance component closest to the chrominance component can be used to perform CCALF on the filtered chrominance component.

As shown in Figure 14, since A and B and their corresponding A1 and B1 are located on different sides of the virtual filter boundary, the luminance component that is located on the same side of the virtual filter boundary as A and B and is closest to the chrominance components A and B is used. A1/B1, perform CCALF on A and B after ALF. The distance refers to the number of pixels, and can also be offset, deviation, or misalignment. The first pixel on the upper side of the virtual filter boundary of the luminance component is taken as the pixel closest to the chrominance components A and B, and the luminance component A1/B1 is used to perform CCALF on A and B after ALF. In this way, when performing CCALF filtering on the image block of the previous row, for the chrominance components A and B, since the SAO of the luminance component A1/B1 has been completed, the chrominance can be achieved in the filtering process of the image block of the previous row. The CCALF of the components A and B, thus completing the CCALF of all the brightness component pixels on the upper side of the virtual filter boundary, without waiting until the brightness component of the next line of image blocks has been SAO, thus eliminating the above-mentioned extra waiting time in CCALF, Thereby improving the filtering efficiency of CCALF and shortening the time of the video encoding and decoding process. At the same time, since there is no waiting process, the buffer does not need to buffer the ALF filter values of A and B, thereby saving the buffer space of the buffer.

In addition to using the luminance component that is located on the same side of the virtual filter boundary as the chrominance component and is closest to the chrominance component for CCALF, you can also use the chrominance component located on the same side of the virtual filter boundary and away from the chrominance component. The non-nearest luminance component performs CCALF on the chrominance component. The non-nearest can be, for example, the second closest, the third closest, etc. The luminance components on the same side of the virtual filter boundary as the chrominance component can be sorted according to the distance from the chrominance component from the nearest to the farthest. The third or fourth luminance component performs CCALF on the chrominance component. For example, in FIG. 14, pixels that are not the closest to the chrominance component A/B (sorted according to the distance from near to far, second or third or fourth) can be used to perform CCALF on the chrominance component A/B. For example, in FIG. 15, the pixel A1/B1 with the second order of the distance to the chrominance component A/B can be used to perform CCALF on the chrominance component A/B.

When there are multiple chrominance components that are respectively located on different sides of the virtual filter boundary and their corresponding luminance components, multiple luminance components closest to the chrominance component can also be used to perform CCALF on the filtered chrominance component. The number of the plurality of luminance component pixels is equal to the number of pixels of the chrominance component that themselves and their corresponding luminance components are located on different sides of the virtual filter boundary.

First, determine the number of pixels of the chrominance component of itself and its corresponding luminance component on different sides of the virtual filter boundary.

Then, the number of luminance components located on the same side of the virtual filter boundary as the chrominance component is determined according to the number of pixels of the chrominance component.

Then determine the brightness component that is closest to the chrominance component and have the same number of pixels as the chrominance component, and use the determined brightness component to perform CCALF on the filtered chrominance component.

For example, in Figure 16, the two chrominance components of themselves and their corresponding brightness components located on different sides of the virtual filter boundary are A and B. It is determined that there are two brightness components located on the same side of the virtual filter boundary as A and B. Determine the two pixels A1 and B1 closest to the chrominance component. A1 and B1 are the second and first pixels located on the upper side of the virtual filtering boundary, respectively. In this embodiment, A1 may be located on the upper side of B1, as shown in FIG. 16; it may also be located on the lower side of B1, as shown in FIG. CCALF is performed on A and B after loop filtering by using A1 and B1 respectively.

The above is only an exemplary description. In addition to the "luminance component closest to the chrominance component" A1/B1 in FIG. 14 can be used, in another example, the luminance component located on the same side of the virtual filter boundary as the chrominance component is used. Performing CCALF on the filtered chrominance components may also include:

The CCALF is performed on the filtered chrominance component using the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component and is a preset distance from the chrominance component.

When there are multiple chrominance components that are located on different sides of the virtual filter boundary with their corresponding luminance components, a luminance component whose distance from the chrominance component is a preset distance can be used to perform CCALF on the filtered chrominance component.

Exemplarily, the preset distance refers to the number of pixels. For example, the preset distance may be two, three, or four pixels. Taking the preset distance of two pixels as an example, as shown in Figure 16, the luminance component A1 that is two pixels away from the chrominance component A and on the same side of the virtual filter boundary is used to perform CCALF on the chrominance component A, and the distance chrominance component is used The luminance component B1 of two pixels B and located on the same side of the virtual filter boundary performs CCALF on the chrominance component B.

When there are multiple chrominance components that are located on different sides of the virtual filter boundary by themselves and their corresponding luminance components, the filtered chrominance components can also be CCALFed by using multiple luminance components at a preset distance from the chrominance component. . The number of the plurality of luminance component pixels is equal to the number of pixels of the chrominance component that themselves and their corresponding luminance components are located on different sides of the virtual filter boundary.

Then determine that the distance to the chrominance component is the brightness component with the same number of pixels of the chrominance component as the preset distance, and use the determined brightness component to perform CCALF on the chrominance component after loop filtering.

For example, in Figure 18, the two chrominance components of themselves and their corresponding luminance components on different sides of the virtual filter boundary are A and B. It is determined that the luminance components to be selected on the same side of the virtual filter boundary as A and B are Two, determine the two pixels A1 and B1 whose distance to the chrominance component is the preset distance. A1 and B1 are the fourth and second pixels located on the upper side of the virtual filter boundary, respectively. A1 and B1 are used to filter the loop respectively CCALF is performed after A and B.

It can be seen that, in this embodiment, when the chrominance component and its corresponding luminance component are located on different sides of the virtual filter boundary, the chrominance component that is located on the same side of the virtual filter boundary as the chrominance component and is closest to the chrominance component or preset The luminance component of the distance filters the chrominance component after loop filtering, which eliminates the extra waiting time in CCALF, improves the filtering efficiency of CCALF, shortens the time of the video encoding and decoding process, and saves the buffer space of the buffer.

Another embodiment of the present disclosure provides a video encoding and decoding method. For the sake of brevity, content that is the same or similar to the previous embodiment will not be repeated. The following only focuses on the content that is different from the previous embodiment.

In this embodiment, prohibiting the use of luminance components located on a different side of the virtual filtering boundary from the chrominance component to perform cross-component filtering on the filtered chrominance component includes:

The chrominance component is prohibited from using cross-component adaptive loop filtering (CCALF).

When the chrominance component and its corresponding luminance component are located on different sides of the virtual filtering boundary, it may be prohibited to use the luminance component corresponding to the chrominance component to perform CCALF on the filtered chrominance component. As shown in Figure 19, the chrominance components A and B are located on the upper side of the virtual filter boundary, while the luminance components A1 and B1 corresponding to the chrominance components A and B are located on the lower side of the virtual filter boundary. In this case, this embodiment CCALF is not performed on chrominance components A and B.

Since the four chrominance component pixels on the upper side of A and B and their corresponding luminance components are all located on the upper side of the virtual filtering boundary, the chrominance components C and D and their corresponding luminance components on the lower side of the virtual filtering boundary are all located on the virtual filtering boundary. On the lower side, CCALF is still performed on the four chrominance components on the upper side of A and B, and the chrominance components C and D. For the aforementioned chrominance component pixels, use the luminance component after SAO to filter the aforementioned chrominance component after ALF, the filter value of CCALF is used as the ALF offset of the chrominance component, and the ALF offset is compared with the The ALF filter values are added, and the result of the addition is used as the loop filter result of the above-mentioned chrominance component. Since CCALF is not performed on A and B, the ALF filter values of A and B are directly used as the loop filter result.

It can be seen that this embodiment prohibits the use of CCALF for the chroma component when the chroma component and its corresponding brightness component are located on different sides of the virtual filtering boundary, and the chroma can be adjusted during the filtering process of the previous line of image blocks. To complete the CCALF of the image block in the previous row, there is no need to wait for the luminance component of the image block in the next row to perform SAO. It can also eliminate the above-mentioned extra waiting time of CCALF, improve the filtering efficiency of CCALF, and shorten the time of the video encoding and decoding process. At the same time, since there is no waiting process, the buffer does not need to buffer the ALF filter values of A and B, thereby saving the buffer space of the buffer.

The above two embodiments described the present disclosure by taking the first direction as the column direction of the luminance component and the chrominance component as an example, but the present disclosure is not limited to this. In other examples, the first direction may also refer to other directions of the luminance component and the chrominance component, such as the row direction. When the first direction is the row direction of the luminance component and the chrominance component, the execution process of the video encoding and decoding method is similar, except that the column direction is replaced with the row direction.

For example, when the first direction is the row direction of the luminance component and the chrominance component, when the luminance component itself and its corresponding luminance component are located on different sides of the virtual filtering boundary, the chrominance component located on the same side of the virtual filtering boundary can be used. The luminance component (for example, the one located on the same side of the virtual filter boundary and the closest distance) performs the CCALF operation on the chrominance component after loop filtering. As shown in FIG. 20, when the first direction is the luminance component and the chrominance component In the row direction, the coding image block, for example, the boundary between the CTUs is perpendicular to the column direction boundary, and the virtual filter boundary is also perpendicular to the column direction virtual filter boundary. For the luminance component, the offset value of the virtual filter boundary can be 4, and any other possible values, the offset value of the virtual filter boundary of the chrominance component can include 2, and any other possible values, and the offset value ratio can be 2: 1, and any other possible values.

In FIG. 20, since A and B are located on the left side of the virtual filter boundary of the chrominance component, and the corresponding luminance component pixel is located on the right side of the virtual filter boundary of the luminance component, the and The luminance component A1/B1 closest to the chrominance components A and B is CCALF performed on A and B after ALF. In this way, when performing CCALF filtering on the image block on the left row, such as CTU, for the chrominance components A and B, since the SAO of the luminance component A1/B1 has been completed, the filtering process of the image block on the left row is sufficient. Realize the CCALF of the chrominance components A and B, thereby completing the CCALF of all the luminance component pixels on the left side of the virtual filter boundary, without waiting for the luminance component of the right row of image blocks to finish SAO, thus eliminating the extra in CCALF Waiting time, thereby improving the filtering efficiency of CCALF and shortening the time of the video encoding and decoding process. At the same time, since there is no waiting process, the buffer does not need to buffer the ALF filter values of A and B, thereby saving the buffer space of the buffer.

Another embodiment of the present disclosure also provides a video encoding and decoding device, as shown in FIG. 21, including:

Memory, used to store executable instructions;

Filter the chrominance components;

When the chrominance component and its corresponding luminance component are on different sides of the virtual filtering boundary, it is forbidden to use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.

The video encoding and decoding device in this embodiment may be a device on the video encoding end, or a device on the video decoding end, or a device having both video encoding and video decoding functions. The processor of this embodiment can execute various operations corresponding to the steps of the foregoing embodiment by executing the executable instructions stored in the memory.

The processor of this embodiment is further configured to perform the following operation: prohibiting the use of luminance components located on a different side of the virtual filtering boundary from the chrominance component to perform cross-component filtering on the filtered chrominance component.

The processor of this embodiment is further configured to perform the following operation: use the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component to perform cross-component filtering on the filtered chrominance component.

The processor of this embodiment is further configured to perform the following operation: use the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component and is closest to the chrominance component to perform cross-component filtering on the filtered chrominance component.

The processor of this embodiment is further configured to perform the following operation: prohibiting the chrominance component from using cross-component adaptive loop filtering.

In this embodiment, the chrominance component and its corresponding brightness component are located on different sides of the virtual filter boundary. It is forbidden to use the brightness component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component. Here, cross-component filtering can be Including cross-component adaptive loop filtering.

The offset value is the offset value of the virtual filter boundary compared to the boundary between the coded image blocks. The coded image block includes a coding tree unit CTU.

The difference between the sampling ratio of the luminance component and the chrominance component in the first direction and the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the first direction includes: the difference between the luminance component and the chrominance component in the first direction The sampling ratio is smaller than the ratio of the offset value of the virtual filter boundary of the luminance component and the chrominance component in the first direction.

The first direction may include columns in the image. The sampling ratio can be 1:1; the offset value ratio can be 2:1.

The case where the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the ratio of the offset value of the virtual filter boundary of the luminance component and the chrominance component in the first direction includes: the image sampling format is 422 or 444 sampling format.

The processor of this embodiment is further configured to perform the following operations: performing any one or more of deblocking filtering, pixel adaptive compensation, or adaptive loop filtering on the chrominance component.

Another embodiment of the present disclosure provides a computer-readable storage medium that stores executable instructions that, when executed by one or more processors, can cause the one or more processors to execute The video encoding and decoding method of the foregoing embodiment.

Another embodiment of the present disclosure provides a video encoding and decoding method. For the sake of brevity, content that is the same as or similar to the foregoing embodiment will not be repeated, and the following only focuses on the content that is different from the foregoing embodiment.

The video encoding and decoding method of this embodiment can be applied to a video encoding end or a video decoding end, and can be applied to a device having both video encoding and video decoding functions, and can solve the problems of low filtering efficiency and large buffer space occupation. As shown in 22, the video encoding and decoding method includes:

Step S2201: filtering the chrominance components.

Step S2202: Use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component, where the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the virtual filtering boundary of the luminance component.

The filtering of the chrominance component in step S2201 includes performing one or more of DF, SAO, or ALF on the reconstructed image block of the chrominance component to obtain a filtering result after the chrominance component is filtered. For example, filtering the chrominance component includes: performing ALF on the reconstructed image block of the chrominance component to obtain the filtering result after ALF.

Exemplarily, the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the filtering boundary of the luminance component, which may include: the sampling ratio of the luminance component and the chrominance component in the first direction and the luminance component and chrominance When the components have different ratios of the offset values of the virtual filter boundary in the first direction, the luminance components corresponding to the chrominance components are used to perform cross-component filtering on the filtered chrominance components.

Exemplarily, for the adaptive loop filtering and cross-component filtering processes of the chrominance component, the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the filtering boundary of the luminance component.

Exemplarily, the correspondence described in this embodiment refers to the correspondence in cross-component adaptive loop filtering, and may also be referred to as association, reference, mapping, and so on. For example, when the image sampling format is 444, the corresponding representation is located in the same position in the chrominance component image block and the luminance component image block.

In this embodiment, the sampling ratio of the luminance component and the chrominance component in the first direction is different from the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the first direction, including: luminance component and chrominance component The sampling ratio in the first direction is smaller than the ratio of the offset value of the virtual filter boundary of the luminance component and the chrominance component in the first direction.

As long as the above sampling ratio is less than the offset value ratio, the offset value of the virtual filter boundary of the chrominance component in this embodiment is equal to the offset value of the virtual filter boundary of the luminance component, and ALF is performed on the chrominance component and the chrominance is used. The luminance component corresponding to the component performs CCALF on the chrominance component.

In the following, the image sampling format is 422 and 444, and the sampling ratio is 1:1 as an example to further explain this embodiment.

As shown in Figure 23, the offset value of the virtual filter boundary of the luminance component in the column direction is 4, and the offset value of the virtual filter boundary of the chrominance component in the column direction is also 4, that is, the luminance component of this embodiment It is the same as the offset value of the virtual filter boundary of the chrominance component. It is equivalent to that the virtual filter boundary of the luminance component shown in FIG. 12 remains unchanged, while the virtual filter boundary of the chrominance component changes from 2 to 4. By keeping the offset values of the virtual filter boundary of the luminance component and the chrominance component consistent, the chrominance components A and B and their corresponding luminance components A1 and B1 in Figure 12 are changed from being on different sides of the virtual filtering boundary to Figure 23 is located on the same side of the virtual filter boundary, that is, on the upper side of the virtual filter boundary, so that the chrominance components A and B and their corresponding luminance components A1 and B1 belong to one line of image blocks, and the lower side of the virtual filter boundary One line of image blocks. In this way, when CCALF filtering is performed on the image block of the previous row, since there is no chrominance component that is on the different side of the virtual filter boundary from the corresponding brightness component as shown in Figure 12, the filtering process of the image block of the previous row can be completed. The CCALF of all the luminance component pixels on the upper side of the virtual filter boundary without waiting for the luminance component of the next row of image blocks to perform SAO. Similarly, when performing CCALF filtering on the next row of image blocks, since there is no corresponding chrominance component that is on the different side of the virtual filter boundary as shown in Figure 12, the filtering process of the next row of image blocks is sufficient. The CCALF of all luminance component pixels on the upper side of the virtual filtering boundary is completed without waiting for the luminance component of the next line of image blocks to perform SAO. In this way, the extra waiting time in CCALF is eliminated, the filtering efficiency of CCALF is improved, and the time of the video encoding and decoding process is shortened. At the same time, since there is no waiting process, the buffer does not need to buffer the ALF filter value of the chrominance component, thereby saving the buffer space of the buffer.

In the above, the offset value of the virtual filter boundary of the luminance component and the chrominance component in the column direction is both 4 as an example, and the present embodiment is described, but the present embodiment is not limited to this. In other implementation manners, the offset value of the virtual filter boundary in the column direction of the luminance component and the chrominance component may both be values greater than 4.

In the above two embodiments, the first direction is the column direction of the luminance component and the chrominance component as an example, and the above embodiments are described, but the present disclosure is not limited to this. In other examples, the first direction may also refer to other directions of the luminance component and the chrominance component, such as the row direction. When the first direction is the row direction of the luminance component and the chrominance component, the execution process of the video encoding and decoding method is similar, except that the column direction is replaced with the row direction.

For example, when the first direction is the row direction of the luminance component and the chrominance component, as shown in FIG. 24, the offset value of the virtual filter boundary of the luminance component in the row direction is 4, and the chrominance component is in the row direction. The offset value of the virtual filter boundary is also 4, and the offset value of the virtual filter boundary through the luminance component and the chrominance component remains the same, so that all the luminance on the left side of the virtual filter boundary can be completed through the filtering process of the left row of image blocks The CCALF of the component pixels does not need to wait until the brightness component of the image block on the right row is SAO, which eliminates the extra waiting time in CCALF, improves the filtering efficiency of CCALF, and shortens the time of the video encoding and decoding process. At the same time, since there is no waiting process, the buffer does not need to buffer the ALF filter value of the chrominance component, thereby saving the buffer space of the buffer.

Another embodiment of the present disclosure also provides a video encoding and decoding device, as shown in FIG. 25, including:

Memory, used to store executable instructions;

Filter the chrominance components;

The luminance component corresponding to the chrominance component is used to perform cross-component filtering on the filtered chrominance component, where the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the virtual filtering boundary of the luminance component.

Exemplarily, the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the filtering boundary of the luminance component, which may include: the sampling ratio of the luminance component and the chrominance component in the first direction and the luminance component and the chrominance component In the case where the ratios of the offset values of the virtual filter boundary in the first direction are different, using the luminance component corresponding to the chrominance component to filter the chrominance component includes: performing CCALF on the chrominance component using the luminance component corresponding to the chrominance component.

In this embodiment, the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component, including: adaptive loop filtering and cross-component filtering processes for the chrominance component, and virtual filtering of the chrominance component The offset value of the boundary is equal to the offset value of the filtering boundary of the luminance component. The offset value is the offset value of the virtual filter boundary compared to the boundary between the coded image blocks. The coded image block includes a coding tree unit CTU.

The difference between the sampling ratio of the luminance component and the chrominance component in the first direction and the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the first direction includes: the difference between the luminance component and the chrominance component in the first direction The sampling ratio is smaller than the ratio of the offset value of the virtual filter boundary of the luminance component and the chrominance component in the first direction. The first direction includes the columns in the image. The sampling ratio can be 1:1; the offset value ratio can be 2:1. The offset value of the virtual filter boundary of the chrominance component and the offset value of the virtual filter boundary of the luminance component may both be 4.

The present disclosure also provides a computer-readable storage medium that stores executable instructions that, when executed by one or more processors, can cause the one or more processors to execute the foregoing embodiments Video encoding and decoding method.

It should be understood that the processor of the embodiment of the present disclosure may be an integrated circuit image processing system with signal processing capability. In the implementation process, the steps of the foregoing method embodiments can be completed by hardware integrated logic circuits in the processor or instructions in the form of software. The above-mentioned processor may be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (ASIC), a ready-made programmable gate array (Field Programmable Gate Array, FPGA) or other Programming logic devices, discrete gates or transistor logic devices, discrete hardware components.

It can be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), and electrically available Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory. The volatile memory may be a random access memory (Random Access Memory, RAM), which is used as an external cache. By way of exemplary but not restrictive description, many forms of RAM are available, such as static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (Double Data Rate SDRAM, DDR SDRAM), Enhanced Synchronous Dynamic Random Access Memory (Enhanced SDRAM, ESDRAM), Synchronous Link Dynamic Random Access Memory (Synchlink DRAM, SLDRAM) ) And Direct Rambus RAM (DR RAM). It should be noted that the memories of the systems and methods described herein are intended to include, but are not limited to, these and any other suitable types of memories.

In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The steps of the method disclosed in combination with the embodiments of the present application may be directly embodied as being executed and completed by a hardware processor, or executed and completed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers. The storage medium is located in the memory, and the processor executes the instructions in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

It should also be understood that the above description of the embodiments of the present application focuses on emphasizing the differences between the various embodiments, and the same or similarities that are not mentioned can be referred to each other, and for the sake of brevity, they will not be repeated here.

The above are only specific implementations of the present disclosure, but the protection scope of the present disclosure is not limited to this. Any person skilled in the art can easily think of various equivalents within the technical scope disclosed in this application. Modifications or replacements, these modifications or replacements shall be covered within the protection scope of this application. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

A video encoding and decoding method, characterized in that it comprises:

Filter the chrominance components;

When the chrominance component and its corresponding luminance component are located on different sides of the virtual filtering boundary, it is prohibited to use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.
The video encoding and decoding method according to claim 1, wherein the prohibiting the use of the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component comprises:

It is forbidden to perform cross-component filtering on the filtered chrominance component using a luminance component located on a different side of the virtual filtering boundary from the chrominance component.
The video encoding and decoding method according to claim 1 or 2, wherein the prohibiting the use of the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component comprises:

Cross-component filtering is performed on the filtered chrominance component using the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component.
The video encoding and decoding method according to claim 3, wherein the chrominance component is used to perform cross-component filtering on the filtered chrominance component using a luminance component that is located on the same side of the virtual filtering boundary as the chrominance component include:

Cross-component filtering is performed on the filtered chrominance component using the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component and is closest to the chrominance component.
The video encoding and decoding method according to claim 1 or 2, wherein the prohibiting the use of the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component comprises:

The chrominance component is prohibited from using cross-component adaptive loop filtering.
The video encoding and decoding method according to any one of claims 1 to 5, wherein when the chrominance component and its corresponding luminance component are located on different sides of the virtual filter boundary, the use of the chrominance component is prohibited Cross-component filtering of the filtered chrominance component by the corresponding luminance component includes:

For the case where the sampling ratio of the luminance component and the chrominance component in the first direction is different from the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the first direction, when the chrominance component and its When the corresponding luminance components are located on different sides of the virtual filtering boundary, it is forbidden to use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.
7. The video encoding and decoding method of claim 6, wherein the offset value is an offset value of the virtual filter boundary compared to the boundary between the coded image blocks.
8. The video coding and decoding method according to claim 7, wherein the coded image block comprises a coding tree unit CTU.
The video encoding and decoding method according to claim 6, wherein the sampling ratio of the luminance component and the chrominance component in the first direction and the virtual filtering boundary of the luminance component and the chrominance component in the first direction The difference in the ratio of the offset value includes that the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the ratio of the offset value of the virtual filtering boundary of the luminance component and the chrominance component in the first direction.
The video encoding and decoding method according to claim 6 or 9, wherein the first direction includes columns in the image.
The video encoding and decoding method of claim 10, wherein the sampling ratio is 1:1; and the offset value ratio is 2:1.
The video encoding and decoding method according to claim 10 or 11, wherein the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the virtual value of the luminance component and the chrominance component in the first direction. The ratio of the offset value of the filter boundary includes that the image sampling format is 422 or 444 sampling format.
The video encoding and decoding method according to any one of claims 1 to 12, wherein the filtering of the chrominance component comprises:

Any one or more of deblocking filtering, pixel adaptive compensation, or adaptive loop filtering is performed on the chrominance component.
A video coding and decoding device, characterized in that it comprises:

Memory, used to store executable instructions;

The processor is configured to execute the executable instructions stored in the memory to perform the following operations:

Filter the chrominance components;

When the chrominance component and its corresponding luminance component are located on different sides of the virtual filtering boundary, it is forbidden to use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.
The video encoding and decoding device according to claim 14, wherein the processor is further configured to perform the following operations:

It is forbidden to perform cross-component filtering on the filtered chrominance component using a luminance component located on a different side of the virtual filtering boundary from the chrominance component.
The video encoding and decoding device according to claim 14 or 15, wherein the processor is further configured to perform the following operations:

Cross-component filtering is performed on the filtered chrominance component using the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component.
The video encoding and decoding device of claim 16, wherein the processor is further configured to perform the following operations:

Cross-component filtering is performed on the filtered chrominance component using the luminance component that is located on the same side of the virtual filtering boundary as the chrominance component and is closest to the chrominance component.
The video encoding and decoding device according to claim 14 or 15, wherein the processor is further configured to perform the following operations:

The chrominance component is prohibited from using cross-component adaptive loop filtering.
The video encoding and decoding device according to any one of claims 14 to 18, wherein the processor is further configured to perform the following operations:

For the case where the sampling ratio of the luminance component and the chrominance component in the first direction is different from the offset value ratio of the virtual filter boundary of the luminance component and the chrominance component in the first direction, when the chrominance component and its When the corresponding luminance components are located on different sides of the virtual filtering boundary, it is forbidden to use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component.
19. The video encoding and decoding device of claim 19, wherein the offset value is an offset value of the virtual filter boundary compared to the boundary between the coded image blocks.
The video coding and decoding device according to claim 20, wherein the coded image block comprises a coding tree unit (CTU).
The video encoding and decoding device according to claim 19, wherein the sampling ratio of the luminance component and the chrominance component in the first direction and the virtual filtering boundary of the luminance component and the chrominance component in the first direction The difference in the ratio of the offset value includes that the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the ratio of the offset value of the virtual filtering boundary of the luminance component and the chrominance component in the first direction.
The video encoding and decoding device according to claim 19 or 22, wherein the first direction includes a column in the image.
The video encoding and decoding device according to claim 23, wherein the sampling ratio is 1:1; and the offset value ratio is 2:1.
The video encoding and decoding device according to claim 23 or 24, wherein the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the virtual value of the luminance component and the chrominance component in the first direction. The ratio of the offset value of the filter boundary includes that the image sampling format is 422 or 444 sampling format.
The video encoding and decoding device according to claim 23 or 24, wherein the processor is further configured to perform the following operations:

Any one or more of deblocking filtering, pixel adaptive compensation, or adaptive loop filtering is performed on the chrominance component.
A computer-readable storage medium, characterized in that it stores executable instructions that, when executed by one or more processors, can cause the one or more processors to execute as claimed in claim 1. The video encoding and decoding method according to any one of claims to 13.
A video encoding and decoding method, characterized in that it comprises:

Filter the chrominance components;

Use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component, wherein the offset value of the virtual filtering boundary of the chrominance component is equal to the offset of the virtual filtering boundary of the luminance component value.
The video encoding and decoding method according to claim 28, wherein the sampling ratio of the luminance component and the chrominance component in the first direction and the virtual filter boundary of the luminance component and the chrominance component in the first direction are determined. In the case of different offset value ratios, the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component.
The video encoding and decoding method according to claim 28 or 29, wherein the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component, including: In the process of adaptive loop filtering and cross-component filtering of the chrominance component, the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the filtering boundary of the luminance component.
The video encoding and decoding method according to any one of claims 28-30, wherein the offset value is an offset value of the virtual filter boundary compared to the boundary between the coded image blocks.
The video coding and decoding method according to claim 31, wherein the coded image block comprises a coding tree unit CTU.
The video encoding and decoding method according to claim 29, wherein the sampling ratio of the luminance component and the chrominance component in the first direction and the virtual filtering boundary of the luminance component and the chrominance component in the first direction The difference in the ratio of the offset value includes that the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the ratio of the offset value of the virtual filtering boundary of the luminance component and the chrominance component in the first direction.
The video encoding and decoding method according to claim 29 or 33, wherein the first direction includes columns in the image.
The video encoding and decoding method of claim 34, wherein the sampling ratio is 1:1; and the offset value ratio is 2:1.
The video encoding and decoding method according to claim 34 or 35, wherein the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the virtual value of the luminance component and the chrominance component in the first direction. The ratio of the offset value of the filter boundary includes that the image sampling format is 422 or 444 sampling format.
The video encoding and decoding method according to any one of claims 28 to 36, wherein the offset value of the virtual filter boundary of the chrominance component and the offset value of the virtual filter boundary of the luminance component are both 4 .
The video encoding and decoding method according to any one of claims 28 to 37, wherein the filtering of the chrominance component comprises:

Any one or more of deblocking filtering, pixel adaptive compensation, or adaptive loop filtering is performed on the chrominance component.
A video coding and decoding device, characterized in that it comprises:

Memory, used to store executable instructions;

The processor is configured to execute the executable instructions stored in the memory to perform the following operations:

Filter the chrominance components;

Use the luminance component corresponding to the chrominance component to perform cross-component filtering on the filtered chrominance component, wherein the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component. Shift value.
The video encoding and decoding device according to claim 39, wherein the sampling ratio of the luminance component and the chrominance component in the first direction and the virtual filter boundary of the luminance component and the chrominance component in the first direction are In the case of different offset value ratios, the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component.
The video encoding and decoding device according to claim 39 or 40, wherein the offset value of the virtual filter boundary of the chrominance component is equal to the offset value of the virtual filter boundary of the luminance component, comprising: In the process of adaptive loop filtering and cross-component filtering of the chrominance component, the offset value of the virtual filtering boundary of the chrominance component is equal to the offset value of the filtering boundary of the luminance component.
The video encoding and decoding device according to any one of claims 39-41, wherein the offset value is an offset value of a virtual filter boundary compared to a boundary between coded image blocks.
The video coding and decoding device according to claim 42, wherein the coded image block comprises a coding tree unit CTU.
The video encoding and decoding device of claim 40, wherein the sampling ratio of the luminance component and the chrominance component in the first direction and the virtual filtering boundary of the luminance component and the chrominance component in the first direction The difference in the ratio of the offset value includes that the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the ratio of the offset value of the virtual filtering boundary of the luminance component and the chrominance component in the first direction.
The video encoding and decoding device according to claim 40 or 44, wherein the first direction includes a column in the image.
The video encoding and decoding device of claim 45, wherein the sampling ratio is 1:1; and the offset value ratio is 2:1.
The video encoding and decoding device according to claim 45 or 46, wherein the sampling ratio of the luminance component and the chrominance component in the first direction is smaller than the virtual value of the luminance component and the chrominance component in the first direction. The ratio of the offset value of the filter boundary includes that the image sampling format is 422 or 444 sampling format.
The video encoding and decoding device according to any one of claims 39 to 47, wherein the offset value of the virtual filter boundary of the chrominance component and the offset value of the virtual filter boundary of the luminance component are both 4 .
The video encoding and decoding device according to any one of claims 39 to 48, wherein the processor is further configured to perform the following operations:

Any one or more of deblocking filtering, pixel adaptive compensation, or adaptive loop filtering is performed on the chrominance component.
A computer-readable storage medium, characterized in that it stores executable instructions that, when executed by one or more processors, can cause the one or more processors to execute as claimed in claim 28 The video encoding and decoding method according to any one of claims to 38.