Intra coding with L-type partitioning tree
AssigneeTencent America LLC
InventorsXin Zhao, Liang Zhao, Shan Liu
Life of this patent
Estimating the term…
Abstract
There is includes a method and apparatus comprising computer code configured to cause a hardware processor or processors to perform obtaining a block of video data, splitting the block into an L-shaped partition and a second partition, the second partition being one of a rectangular shaped partition and another L-shaped partition, and performing intra prediction of the L-shaped partition by using a reference sample chain, comprising a chain of neighboring reconstructed samples, as reference samples.
Claims20
A method of video decoding performed by at least one processor, comprising:
- receiving, via a video bitstream, coded video data comprising a block of video data;
- partitioning the block into at least an L-shaped partition and a second partition;
- performing an intra prediction for the L-shaped partition by using a reference sample chain, comprising a horizontal reference line and a vertical reference line, that spans a boundary between the L-shaped partition and the second partition;
- generating a predictor for the second partition by using the reference sample chain that spans the boundary between the L-shaped partition and the second partition;
- and
- reconstructing the block based on the intra prediction of the L-shaped partition and the predictor for the second partition.
A method of video encoding performed by at least one processor, comprising:
- receiving video data comprising a block of video data;
- partitioning the block into at least an L-shaped partition and a second partition;
- performing an intra prediction for the L-shaped partition by using a reference sample chain, the reference sample chain comprising a horizontal reference line and a vertical reference line, wherein the reference sample chain spans a boundary between the L-shaped partition and the second partition;
- generating a predictor for the second partition by using the reference sample chain that spans the boundary between the L-shaped partition and the second partition;
- and
- encoding the block based on the intra prediction of the L-shaped partition and the predictor for the second partition.
A method of processing visual media data performed by at least one processor, comprising:
- obtaining a source video sequence;
- and
- performing a conversion between the source video sequence and a bitstream of visual media data, wherein:
- the bitstream comprises encoded video data partitioned into a plurality of blocks, including a first block that has been encoded at least in part using an intra prediction and a predictor, the intra prediction corresponding to an L-shaped partition of the first block and a reference sample chain for the first block, and the predictor corresponding to a second partition of the first block and the reference sample chain.
Dependencies are inferred from the claim text.