Combined coding of split information and other block-level parameters for video coding/decoding
A block of size N×M of a video frame can be either split into smaller blocks in n different ways (e.g. horizontal, vertical, quad, etc.), or alternatively it can be coded using one out of m different coding modes (e.g. intra, inter, bi-prediction, etc.). Since splitting into smaller sub-blocks and coding the N×M block in one of the m possible ways are mutually exclusive, there are n+m distinct possibilities for processing the block further. The n+m possible ways of processing the N×M blocks are assigned a code (e.g., a number) and coded with a codeword derived from a VLC table. This achieves efficient combined coding of split information and other block-level parameters.
1. A method comprising:
partitioning a video frame into a plurality of blocks;
for each of the plurality of blocks, either splitting the block into sub-blocks using one of n different partitioning modes or coding the block using one of m different coding modes such that there are n+m processing modes for each block; and
encoding a given block with a codeword representing one of the n+m processing modes for use in deriving an encoded bit-stream, wherein encoding includes using different variable length codes depending on information from blocks to the left and above the given block.
2. The method of claim 1 , wherein the information from blocks to the left and above the given block includes information indicating a skip mode or no skip mode.
3. The method of claim 2 , wherein the different variable length codes for the different coding modes include:
SPLIT
1
INTER0
01
SKIP
001
MERGE
0001
INTRA
00001
INTER1
000001
INTER2
0000001
INTER3
00000001
BIPRED
00000000.
4. The method of claim 1 , wherein the n partitioning modes include partitioning symmetrically into N×M sub-blocks where all sub-blocks have the same size horizontally, and all sub-blocks have the same size vertically.
5. The method of claim 1 , wherein the n partitioning modes include partitioning asymmetrically into N×M sub-blocks where at least two sub-blocks have different size either horizontally or vertically.
6. The method of claim 1 , wherein the n partitioning modes include a partitioning mode in which the block is first split horizontally and then at least one resulting sub-block is split vertically, or a partitioning mode in which the block is first split vertically and then at least one resulting sub-block is split horizontally.
7. The method of claim 1 , wherein the m coding modes include intra prediction, skip, merge, inter prediction with uni-prediction, and inter prediction with bi-prediction.
8. The method of claim 7 , wherein the inter prediction with uni-prediction mode includes several sub-modes that depend on a reference index.
9. A method comprising:
receiving an encoded bit-stream representing video frames to be decoded;
deriving from the encoded bit-stream for each of a plurality of blocks partitioned from a given video frame, a codeword representing a processing mode, among a plurality of processing modes, for each block, wherein the codeword for a given block is derived from different variable length codes depending on information from blocks to the left and above the given block;
determining from the codeword which one of n different partitioning modes is used for splitting the given block into sub-blocks or which one of m different coding modes is used for the given block of the plurality of blocks of the given video frame; and
decoding the given block of the given video frame based on the which one of n different partitioning modes or which one of m different coding modes is used for the given block of the plurality of blocks of the given video frame according to the codeword.
10. The method of claim 9 , wherein the information from blocks to the left and above the given block includes information indicating a skip mode or no skip mode.
11. The method of claim 10 , wherein the different variable length codes include:
SPLIT
1
INTER0
01
SKIP
001
MERGE
0001
INTRA
00001
INTER1
000001
INTER2
0000001
INTER3
00000001
BIPRED
00000000.
12. The method of claim 9 , wherein the n partitioning modes include partitioning symmetrically into N×M sub-blocks where all sub-blocks have the same size horizontally, and all sub-blocks have the same size vertically.
13. The method of claim 9 , wherein the n partitioning modes include partitioning asymmetrically into N×M sub-blocks where at least two sub-blocks have different size either horizontally or vertically.
14. The method of claim 9 , wherein the n partitioning modes include a partitioning mode in which the block is first split horizontally and then at least one resulting sub-block is split vertically, or a partitioning mode in which the block is first split vertically and then at least one resulting sub-block is split horizontally.
15. The method of claim 9 , wherein the m coding modes include intra prediction, skip, merge, inter prediction with uni-prediction, and inter prediction with bi-prediction.
16. The method of claim 15 , wherein the inter prediction with uni-prediction mode includes several sub-modes that depend on a reference index.
17. One or more non-transitory computer readable storage media encoded with executable instructions that, when executed by a processor, cause the processor to perform operations comprising:
receiving an encoded bit-stream representing video frames to be decoded;
deriving from the encoded bit-stream for each of a plurality of blocks partitioned from a given video frame, a codeword representing a processing mode, among a plurality of processing modes, for each block, wherein the codeword for a given block is derived from different variable length codes depending on information from blocks to the left and above the given block;
determining from the codeword which one of n different partitioning modes is used for splitting the given block into sub-blocks or which one of m different coding modes is used for the given block of the plurality of blocks of the given video frame; and
decoding the given block of the given video frame based on the which one of n different partitioning modes or which one of m different coding modes is used for the given block of the plurality of blocks of the given video frame according to the codeword.
18. The non-transitory computer readable storage media of claim 17 , wherein the information from blocks to the left and above the given block includes information indicating a skip mode or no skip mode.
19. The non-transitory computer readable storage media of claim 18 , wherein the different variable length codes include:
SPLIT
1
INTER0
01
SKIP
001
MERGE
0001
INTRA
00001
INTER1
000001
INTER2
0000001
INTER3
00000001
BIPRED
00000000.
20. The non-transitory computer readable storage media of claim 17 , wherein the n partitioning modes include partitioning symmetrically into N×M sub-blocks where all sub-blocks have the same size horizontally, and all sub-blocks have the same size vertically.
21. The non-transitory computer readable storage media of claim 17 , wherein the n partitioning modes include partitioning asymmetrically into N×M sub-blocks where at least two sub-blocks have different size either horizontally or vertically.
22. The non-transitory computer readable storage media of claim 17 , wherein the n partitioning modes include a partitioning mode in which the block is first split horizontally and then at least one resulting sub-block is split vertically, or a partitioning mode in which the block is first split vertically and then at least one resulting sub-block is split horizontally.
23. The non-transitory computer readable storage media of claim 17 , wherein the m coding modes include intra prediction, skip, merge, inter prediction with uni-prediction, and inter prediction with bi-prediction.
24. The non-transitory computer readable storage media of claim 23 , wherein the inter prediction with uni-prediction mode includes several sub-modes that depend on a reference index.