Context modeling constraints for block differential pulse-code modulation
A method of video encoding performed in a video encoder is provided. In the method, context modeling is performed to determine a context model for each of a number of bins of syntax elements corresponding to residues of a region of a transform skipped block in a coded picture. The transform skipped block is coded with Block Differential Pulse-code Modulation (BDPCM) and divided into a plurality of regions. The number of the bins of syntax elements being context coded does not exceed a maximum number of context coded bins set for the region. The maximum number of context coded bins is determined based on a comparison between a threshold and a number of quantized residues in the region. A bit stream including coded bits of the bins of syntax elements is generated based on the determined context models.
1 . A method of video decoding performed in a video decoder, the method comprising:
receiving a bitstream including bins of syntax elements, the syntax elements corresponding to residuals of a plurality of regions of a transform skip block in an encoded image;
performing context modeling to determine a context model for each of a number of the bins of the syntax elements of a region of the plurality of regions, the number of the bins of the syntax elements of the region of the transform skip block that is context coded not exceeding a maximum number of context-coded bins set for the region of the plurality of regions of the transform skip block; and
decoding the number of the bins of the syntax elements based on the determined context model.
2 . The method of claim 1 , wherein the maximum number of context-coded bins set for the region is less than or equal to a number of the bins of the syntax elements of the region of the transform skip block.
3 . The method of claim 1 , wherein the maximum number of the context coded bins set for the region of the transform skip block is B×A, B being a positive number and A being a number of quantized residuals in the region.
4 . The method of claim 3 , wherein B is an integer.
5 . The method of claim 3 , wherein B is a fraction.
6 . The method of claim 3 , wherein B is set according to A.
7 . The method of claim 3 , wherein
the transform skip block is divided into the plurality of regions including the region;
the maximum number of context coded bins in each region of the plurality of regions of the transform skip block is B×A, B being the positive number and A being a number of quantized residuals in the respective region of the transform skip block.
8 . The method of claim 1 , further comprising:
decoding coded bits of a remaining number of the bins of the syntax elements that is not context coded for the region of the transform skip block based on an equal probability model.
9 . The method of claim 1 , wherein the transform skip block is associated with a current block that is coded with differential pulse-code modulation (BDPCM).
10 . A method of video encoding performed in a video encoder, the method comprising:
obtaining bins of syntax elements, the syntax elements corresponding to residuals of a plurality of regions of a transform skip block in an encoded image;
performing context modeling to determine a context model for each of a number of the bins of the syntax elements of a region of the plurality of regions, the number of the bins of the syntax elements of the region of the transform skip block that is context coded not exceeding a maximum number of context-coded bins set for the region of the plurality of regions of the transform skip block; and
encoding the number of the bins of the syntax elements based on the determined context model.
11 . The method of claim 10 , wherein the maximum number of context-coded bins set for the region is less than or equal to a number of the bins of the syntax elements of the region of the transform skip block.
12 . The method of claim 10 , wherein the maximum number of the context coded bins set for the region of the transform skip block is B×A, B being a positive number and A being a number of quantized residuals in the region.
13 . The method of claim 12 , wherein B is an integer.
14 . The method of claim 12 , wherein B is a fraction.
15 . The method of claim 12 , wherein B is set according to A.
16 . The method of claim 12 , wherein the transform skip block is divided into the plurality of regions including the region;
the maximum number of context coded bins in each region of the plurality of regions of the transform skip block is B×A, B being the positive number and A being a number of quantized residuals in the respective region of the transform skip block.
17 . The method of claim 10 , further comprising:
encoding a remaining number of bins of the syntax elements that is not context coded for the region of the transform skip block based on an equal probability model.
18 . The method of claim 10 , wherein the transform skip block is associated with a current block that is coded with differential pulse-code modulation (BDPCM).
19 . A non-transitory computer-readable storage medium storing instructions which when executed by a processor cause the processor to perform a video encoding method, the video encoding method comprising:
obtaining bins of syntax elements, the syntax elements corresponding to residuals of a plurality of regions of a transform skip block in an encoded image;
performing context modeling to determine a context model for each of a number of the bins of the syntax elements of a region of the plurality of regions, the number of the bins of the syntax elements of the region of the transform skip block that is context coded not exceeding a maximum number of context-coded bins set for the region of the plurality of regions of the transform skip block;
encoding the number of the bins of the syntax elements in a video bitstream based on the determined context model; and
transmitting the video bitstream.
20 . The non-transitory computer-readable storage medium of claim 19 , wherein the maximum number of context-coded bins set for the region is less than or equal to a number of the bins of the syntax elements of the region of the transform skip block.