Method and apparatus for video coding
Aspects of the disclosure provide methods and apparatuses for video encoding/decoding. In some examples, an apparatus for video decoding includes receiving circuitry and processing circuitry. The processing circuitry decodes partition information of a block in a picture from a coded video bitstream. The partition information is indicative of a partition tree structure for a partition of the block into coding blocks. Then, the processing circuitry groups to-be reconstructed samples of the block into processing groups according to the partition of the block into the coding blocks. A shape of a combined region of a processing group is limited by constraints in the partition of the block into the coding blocks. Then, the processing circuitry reconstructs the to-be reconstructed samples of the block by multiple pipeline stages that process the processing groups in a pipeline processing manner.
1. A method for video decoding in a decoder, comprising:
decoding partition information of a block in a current picture from a coded video bitstream, the partition information being indicative of a partition tree structure for partitioning of the block into coding blocks;
grouping to-be reconstructed samples of the block into processing groups according to the coding blocks, each one of the processing groups corresponding to a respective region of a same number of samples having a shape that is limited by constraints in the partitioning of the block into the coding blocks, and the constraints including
allowing a ternary tree (TT) split of a non-leaf partition when the non-leaf partition is at a coding tree unit (CTU) level,
allowing the TT split of the non-leaf partition when the non-leaf partition is below the CTU level and contained in a predetermined area, and
prohibiting the TT split of the non-leaf partition when the non-leaf partition is below the CTU level and not contained in the predetermined area; and
reconstructing the to-be reconstructed samples of the block by multiple pipeline stages that process the processing groups in a pipeline processing manner.
2. The method of claim 1 , wherein the shape of the region of one of the processing groups is limited to be one of a square shape and a rectangular shape.
3. The method of claim 1 , wherein the shape of the region of one of the processing groups has no more than a fixed number of edges or has no more than a fixed number of vertices.
4. The method of claim 1 , wherein the constraints further include no more than a fixed number of concatenating TT splits in the partition tree structure for the partitioning of the block into the coding blocks.
5. The method of claim 1 , wherein the region of one of the processing groups has at least one of a width or a height equal to or larger than a fixed number.
6. The method of claim 1 , wherein one of the processing groups includes samples from multiple coding blocks, and a number of the multiple coding blocks is constrained to be no more than a limit value.
7. The method of claim 6 , wherein the limit value is determined based on a size of the region of the one of the processing groups.
8. The method of claim 1 , wherein the constraints further include limiting usages of a TT split and a binary tree split in the partition tree structure when the current picture is a non-intra picture.
9. The method of claim 1 , wherein the region of one of the processing groups includes a portion that is outside a boundary of the current picture.
10. An apparatus for video decoding, comprising:
processing circuitry configured to:
decode partition information of a block in a current picture from a coded video bitstream, the partition information being indicative of a partition tree structure for partitioning of the block into coding blocks;
group to-be reconstructed samples of the block into processing groups according to the coding blocks, each one of the processing groups corresponding to a respective region of a same number of samples having a shape that is limited by constraints in the partitioning of the block into the coding blocks, and the constraints including
allowing a ternary tree (TT) split of a non-leaf partition when the non-leaf partition is at a coding tree unit (CTU) level,
allowing the TT split of the non-leaf partition when the non-leaf partition is below the CTU level and contained in a predetermined area, and
prohibiting the TT split of the non-leaf partition when the non-leaf partition is below the CTU level and not contained in the predetermined area: and
reconstruct the to-be reconstructed samples of the block by multiple pipeline stages that process the processing groups in a pipeline processing manner.
11. The apparatus of claim 10 , wherein the shape of the region of one of the processing groups is limited to be one of a square shape and a rectangular shape.
12. The apparatus of claim 10 , wherein the shape of the region of one of the processing groups has no more than a fixed number of edges or has no more than a fixed number of vertices.
13. The apparatus of claim 10 , wherein the constraints further include no more than a fixed number of concatenating TT splits in the partition tree structure for the partitioning of the block into the coding blocks.
14. The apparatus of claim 10 , wherein the region of one of the processing groups has at least one of a width or a height equal to or larger than a fixed number.
15. The apparatus of claim 10 , wherein one of the processing groups includes samples from multiple coding blocks, and a number of the multiple coding blocks is constrained to be no more than a limit value.
16. The apparatus of claim 15 , wherein the limit value is determined based on a size of the region of the one of the processing groups.
17. The apparatus of claim 10 , wherein the constraints further include limiting usages of a TT split and a binary tree split in the partition tree structure when the current picture is a non-intra picture.
18. A non-transitory computer-readable medium storing instructions which when executed by a computer for video decoding cause the computer to perform:
decoding partition information of a block in a current picture from a coded video bitstream, the partition information being indicative of a partition tree structure for partitioning of the block into coding blocks;
grouping to-be reconstructed samples of the block into processing groups according to the coding blocks, each one of the processing groups corresponding to a respective region of a same number of samples having a shape that is limited by constraints in the partitioning of the block into the coding blocks, and the constraints including
allowing a ternary tree (TT) split of a non-leaf partition when the non-leaf partition is at a coding tree unit (CTU) level,
allowing the TT split of the non-leaf partition when the non-leaf partition is below the CTU level and contained in a predetermined area, and
prohibiting the TT split of the non-leaf partition when the non-leaf partition is below the CTU level and not contained in the predetermined area; and
reconstructing the to-be reconstructed samples of the block by multiple pipeline stages that process the processing groups in a pipeline processing manner.
19. The non-transitory computer-readable medium of claim 18 , wherein the shape of the region of one of the processing groups has no more than a fixed number of edges or has no more than a fixed number of vertices.
20. The non-transitory computer-readable medium of claim 18 , wherein the constraints further include no more than a fixed number of concatenating TT splits in the partition tree structure for the partitioning the block into the coding blocks.