Granted Patent B2

US 12,388,979 · App. 17/974,068 · Granted Aug 12, 2025

Inventors: Xin Zhao (San Jose, CA); Xiaozhong Xu (State College, PA); Shan Liu (San Jose, CA)

Assignee: Tencent America LLC

H04N19/105H04N19/159H04N19/176H04N19/46

Abstract

Claims (66)

1. A method for processing video data, the method comprising:

receiving a video bitstream comprising a current block of a video frame;

extracting an Intra-Block Copy (IntraBC) flag from the video bitstream indicating that the current block is predicted under an IntraBC mode;

determining from the video bitstream that the current block is partitioned under a wedgelet partitioning mode, wherein the current block is partitioned under the wedgelet partitioning mode into a plurality of partitions comprising a first partition and a second partition;

identifying at least the first partition and the second partition of the current block;

generating a list of candidate block vectors based on at least one of: block vectors from neighboring blocks of the current block, or historical block vectors, the list of candidate block vectors comprising at least one candidate block vector;

extracting, from the video bitstream, a third indicator indicating a target block vector in the list of candidate block vectors;

selecting the target block vector from the list of candidate block vectors according to the third indicator;

generating, based on the target block vector, at least one of: a first block vector for predicting the first partition in the IntraBC mode; or a second block vector for predicting the second partition in the IntraBC mode; and

decoding the current block based on at least the first block vector and second block vector.

2. The method of claim 1 , wherein determining from the video bitstream that the current block is partitioned under the wedgelet partitioning mode comprises:

extracting, from the video bitstream, a first indicator for indicating the wedgelet partitioning mode for the current block; and

determining that the current block is partitioned under the wedgelet partitioning mode based on the wedgelet partitioning mode indicator.

3. The method of claim 1 , wherein:

the method further comprises extracting, from the video bitstream, a second indicator associated with the current block and indicating a pattern of the wedgelet partitioning mode; and

identifying the first partition and the second partition of the current block comprises identifying the first partition and the second partition based on the second indicator.

4. The method of claim 3 , wherein the pattern of the wedgelet partitioning mode comprises one of:

a vertical partitioning pattern in which the first partition and the second partition are divided by a vertical boundary in the current block; or

a horizontal partitioning pattern in which the first partition and the second partition are divided by a horizontal boundary.

5. The method of claim 3 , wherein extracting, from the video bitstream, the second indicator comprises:

extracting the second indicator from the video bitstream via a high-level syntax, the high-level syntax comprising at least one of:

a Video Parameter Set (VPS) syntax;

a Picture Parameter Set (PPS) syntax;

a Sequence Parameter Set (SPS) syntax;

an Adaptive Parameter Set (APS) syntax;

a slice header;

a picture header;

a frame header; or

a tile header.

6. The method of claim 3 , wherein:

extracting, from the video bitstream, the second indicator comprises extracting the second indicator via a block-level signal from the video bitstream;

the block-level signal is transmitted in one of:

a Coding Tree Unit (CTU) header;

s superblock header; or

a coding block header; and

the second indicator applies to one of:

a CTU comprising the current block in response to the block-level signal being transmitted in the CTU header;

a superblock comprising the current block in response to the block-level signal being transmitted in the superblock header; or

a coding block comprising the current block in response to the block-level signal being transmitted in the coding block header.

7. The method of claim 1 , wherein the first block vector and second block vector are different.

8. The method of claim 7 , wherein extracting the second block vector for the second partition comprises predicting the second block vector based on the first block vector.

9. The method of claim 1 , wherein generating at least one of the first block vector or the second block vector based on the target block vector comprises:

extracting, from the video bitstream, vector prediction residues associated with the first block vector; and

generating the first block vector based on the target block vector and the vector prediction residues.

10. The method of claim 1 , wherein generating at least one of the first block vector or the second block vector based on the target block vector comprises deriving the second block vector from the target block vector.

11. The method of claim 10 , wherein deriving the second block vector from the target block vector comprises generating the second block vector from the target block vector using a predefined transformation.

12. The method of claim 1 , wherein the first block vector points to an IntraBC prediction block in a current frame of the current block, and the IntraBC prediction block is determined by an encoder of the video bitstream by:

determining a first search area in the video frame, wherein the first search area is a first candidate area for locating the IntraBC prediction block, wherein the first search area has no overlap with the current block and comprises a list of blocks, the IntraBC prediction block being a prediction block for performing IntraBC prediction on the first partition;

determining a second search area, wherein the second search area is a second candidate area for locating the IntraBC prediction block, the second search area comprising at least one of: (i) a sub-block of the current block, and (ii) an adjacent block of the current block; and

identifying the IntraBC prediction block pointed to by the first block vector which does not cross search areas.

13. The method of claim 12 , wherein:

a top left pixel of the current block has a coordinate position of (x0, y0);

a top left pixel of each block in the list of blocks has a coordinate position of (x, y);

y is less than y0; and

x is less than [x0+2 (y0−y)−D],

where x0, y0, x, and y are non-negative numbers, D is a number of immediate reconstructed blocks that is restricted for IntraBC mode.

14. A device for processing video data, the device comprising a memory for storing computer instructions and a processor in communication with the memory, wherein, when the processor executes the computer instructions, the processor is configured to cause the device to:

receive a video bitstream comprising a current block of a video frame;

extract an Intra-Block Copy (IntraBC) flag from the video bitstream indicating that the current block is predicted under an IntraBC mode;

determine from the video bitstream that the current block is partitioned under a wedgelet partitioning mode, wherein the current block is partitioned under the wedgelet partitioning mode into a plurality of partitions comprising a first partition and a second partition;

identify at least the first partition and the second partition of the current block;

generate a list of candidate block vectors based on at least one of: block vectors from neighboring blocks of the current block, or historical block vectors, the list of candidate block vectors comprising at least one candidate block vector;

extract, from the video bitstream, a third indicator indicating a target block vector in the list of candidate block vectors;

select the target block vector from the list of candidate block vectors according to the third indicator;

generate, based on the target block vector, at least one of: a first block vector for predicting the first partition in the IntraBC mode; or a second block vector for predicting the second partition in the IntraBC mode; and

decode the current block based on at least the first block vector and second block vector.

Assignments (1)

Continuity (3)

References Cited (11)