Encoding method and apparatus
A video processing method includes: obtaining prediction information of a CU; obtaining, when the CU comprises only one residual TU and a size of the residual TU is less than a size of the CU, a TU partitioning mode of the CU and a residual position of the residual TU, wherein the TU partitioning mode and the residual position are used to determine a horizontal transform type and a vertical transform type; obtaining transform coefficients of the residual TU based on the horizontal transform type and the vertical transform type; and generating a bitstream that is for storing or transmitting and that includes the prediction information, a first flag that indicates the TU partitioning mode, a second flag that indicates the residual position, and the transform coefficients.
1 . A video encoding method, comprising:
obtaining a picture;
partitioning, based on partition information, the picture into a set of non-overlapping blocks that include a coding unit (CU);
obtaining, when the CU comprises only one residual transform unit (TU) and when a first size of the residual TU is less than a second size of the CU, a TU partitioning mode of the CU and a residual position of the residual TU;
determining, using the TU partitioning mode and the residual position, a horizontal transform type and a vertical transform type;
obtaining transform coefficients of the residual TU based on the horizontal transform type and the vertical transform type; and
generating a bitstream that comprises a first flag indicating the TU partitioning mode, a second flag indicating the residual position, the partition information, and the transform coefficients.
2 . The video encoding method of claim 1 , wherein the horizontal transform type is based on discrete sine transform (DST) type VII (DST-7) and the vertical transform type is based on the DST-7, the horizontal transform type is based on the DST-7 and the vertical transform type is based on discrete cosine transform (DCT) type VIII (DCT-8), or the horizontal transform type is based on the DCT-8 and the vertical transform type is based on the DST-7.
3 . The video encoding method of claim 2 , wherein the TU partitioning mode specifies that the CU is partitioned into two first TUs that are arranged horizontally and that comprise a left TU and a right TU, and a third size of the left TU is smaller than, greater than, or equal to a fourth size of the right TU, or wherein the TU partitioning mode specifies that the CU is partitioned into two second TUs that are arranged vertically and that comprise a top TU and a bottom TU, and a fifth size of the top TU is smaller than, greater than, or equal to a sixth size of the bottom TU.
4 . The video encoding method of claim 3 , further comprising:
partitioning the CU into the two first TUs, wherein the residual position is left or right; or
partitioning the CU into the two second TUs, wherein the residual position is a top or a bottom.
5 . The video encoding method of claim 4 , wherein the horizontal transform type is the DST-7 and the vertical transform type is the DCT-8 when the CU is partitioned into the two second TUs and the residual position is the top, the horizontal transform type is the DST-7 and the vertical transform type is the DST-7 when the CU is partitioned into the two second TUs and the residual position is the bottom, the horizontal transform type is the DCT-8 and the vertical transform type is the DST-7 when the CU is partitioned into the two first TUs and the residual position is the left, and the horizontal transform type is the DST-7 and the vertical transform type is the DST-7 when the CU is partitioned into the two first TUs and the residual position is the right.
6 . The video encoding method of claim 1 , wherein the residual TU comprises one or more residuals for one or more of a first chrominance component, a second chrominance component, or a luminance component.
7 . The video encoding method of claim 6 , wherein the one or more residuals comprise a first residual for the first chrominance component, a second residual for the second chrominance component, and a third residual for the luminance component.
8 . A video decoding method, comprising:
obtaining partition information;
partitioning, based on the partition information, a picture block into a set of non-overlapping blocks that include a coding unit (CU);
obtaining, when the CU comprises only one residual transform unit (TU) and a size of the residual TU is less than a size of the CU, a TU partitioning mode of the CU and a residual position of the residual TU;
obtaining, based on the TU partitioning mode and the residual position, a horizontal transform type and a vertical transform type;
applying, based on the horizontal transform type and the vertical transform type, an inverse transform to dequantized coefficients of the residual TU to obtain a residual block of the residual TU; and
obtaining a video block based on the residual block.
9 . The video decoding method of claim 8 , wherein the horizontal transform type is based on discrete sine transform (DST) type VII (DST-7) and the vertical transform type is based on the DST-7, the horizontal transform type is based on the DST-7 and the vertical transform type is based on discrete cosine transform (DCT) type VIII (DCT-8), or the horizontal transform type is based on the DCT-8 and the vertical transform type is based on the DST-7.
10 . The video decoding method of claim 9 , wherein the TU partitioning mode specifies that the CU is partitioned into two first TUs that are arranged horizontally and that comprise a left TU and a right TU, and a third size of the left TU is smaller than, greater than, or equal to a fourth size of the right TU, or wherein the TU partitioning mode specifies that the CU is partitioned into two second TUs that are arranged vertically and that comprise a top TU and a bottom TU, and a fifth size of the top TU is smaller than, greater than, or equal to a sixth size of the bottom TU.
11 . The video decoding method of claim 10 , further comprising:
partitioning the CU into the two first TUs, wherein the residual position is left or right; or
partitioning the CU into the two second TUs, wherein the residual position is a top or a bottom.
12 . The video decoding method of claim 11 , wherein the horizontal transform type is the DST-7 and the vertical transform type is the DCT-8 when the CU is partitioned into the two second TUs and the residual position is the top, the horizontal transform type is the DST-7 and the vertical transform type is the DST-7 when the CU is partitioned into the two second TUs and the residual position is the bottom, the horizontal transform type is the DCT-8 and the vertical transform type is the DST-7 when the CU is partitioned into the two first TUs and the residual position is the left, and the horizontal transform type is the DST-7 and the vertical transform type is the DST-7 when the CU is partitioned into the two first TUs and the residual position is the right.
13 . The video decoding method of claim 8 , wherein the residual TU comprises one or more residuals for one or more of a first chrominance component, a second chrominance component, or a luminance component.
14 . The video decoding method of claim 13 , wherein the one or more residuals comprise a first residual for the first chrominance component, a second residual for the second chrominance component, and a third residual for the luminance component.
15 . A non-transitory computer-readable storage medium storing a bitstream and one or more instructions executable by at least one processor to perform operations of decoding the bitstream, the operations comprising:
receiving the bitstream, the bitstream comprising:
partition information configured to partition a picture block into a set of non-overlapping blocks that includes a coding unit (CU);
a first flag indicating a transform unit (TU) partitioning mode that partitions the CU into at least two non-overlapping TUs that include a residual TU;
transform coefficients of the residual TU; and
a second flag indicating a residual position of the residual TU,
wherein the TU partitioning mode and the residual position determine a horizontal transform type and a vertical transform type,
wherein the horizontal transform type, the vertical transform type, and the transform coefficients obtain a residual block of the residual TU, and
wherein the residual block obtains a video block.
16 . The non-transitory computer-readable medium of claim 15 , wherein the horizontal transform type is based on discrete sine transform (DST) type VII (DST-7) and the vertical transform type is based on the DST-7, the horizontal transform type is based on the DST-7 and the vertical transform type is based on discrete cosine transform (DCT) type VIII (DCT-8), or the horizontal transform type is based on the DCT-8 and the vertical transform type is based on the DST-7.
17 . The non-transitory computer-readable medium of claim 16 , wherein the TU partitioning mode specifies that the CU is partitioned into two first TUs that are arranged horizontally and that comprise a left TU and a right TU, and a third size of the left TU is smaller than, greater than, or equal to a fourth size of the right TU, or wherein the TU partitioning mode specifies that the CU is partitioned into two second TUs that are arranged vertically and that comprise a top TU and a bottom TU, and a fifth size of the top TU is smaller than, greater than, or equal to a sixth size of the bottom TU.
18 . The non-transitory computer-readable medium of claim 17 , wherein the residual position is either left or right when the CU is partitioned into the two first TUs, and wherein the residual position is either a top or a bottom when the CU is partitioned into the two second TUs.
19 . The non-transitory computer-readable medium of claim 18 , wherein the horizontal transform type is the DST-7 and the vertical transform type is the DCT-8 when the CU is partitioned into the two second TUs and the residual position is the top, the horizontal transform type is the DST-7 and the vertical transform type is the DST-7 when the CU is partitioned into the two second TUs and the residual position is the bottom, the horizontal transform type is the DCT-8 and the vertical transform type is the DST-7 when the CU is partitioned into the two first TUs and the residual position is the left, and the horizontal transform type is the DST-7 and the vertical transform type is the DST-7 when the CU is partitioned into the two first TUs and the residual position is the right.
20 . The non-transitory computer-readable medium of claim 15 , wherein the residual TU comprises one or more residuals for one or more of a first chrominance component, a second chrominance component, or a luminance component.