IP Library › Granted Patent US 12701220

Granted Patent B2

US 12701220 · App. 18/913,797 · Granted Aug 4, 2026

Method, apparatus, and medium for video processing

Inventors: Zhipin Deng (Beijing, CN); Kai Zhang (Los Angeles, CA); Li Zhang (Los Angeles, CA)

Assignees: BEIJING BYTEDANCE NETWORK TECHNOLOGY CO., LTD.; BYTEDANCE INC.

H04N19/105 H04N19/176

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 12701220
App. No.: 18/913,797
Granted: Aug 4, 2026
Kind: B2

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. The method comprises: determining, for a conversion between a current video block of a video and a bitstream of the video, motion information of the current video block, the current video block being coded with at least one of: an intra block copy (IBC) merge mode, an IBC with template matching mode, or an intra template matching mode; updating the motion information based on a constraint, the constraint indicating a target value of a component of the motion information; and performing the conversion based on the updated motion information.

Claims (125)

1 . A method for video processing, comprising:

determining, for a conversion between a current video block of a video and a bitstream of the video, a reference template of the current video block based on coding information of the current video block, the current video block being coded with a sample reordering mode; and

performing the conversion based on the reference template.

2 . The method of claim 1 , wherein the coding information comprises at least one of:

a sample reordering type of the sample reordering mode, or

a template shape of the reference template.

3 . The method of claim 1 , further comprising: determining motion information of the current video block based on the coding information of the current video block,

wherein the motion information comprises one of: a motion vector of the current video block, or a block vector of the current video block,

wherein the coding information comprises at least one of: a sample reordering type of the sample reordering mode, a template shape of the reference template, a dimension of the current video block, a dimension of a template of the current video block, a dimension of a partial of the template, a location of the current video block, or a location of the template,

wherein the template of the current video block comprises at least one of: a current template of the current video block, or the reference template of the current video block,

wherein the dimension of the current video block comprises at least one of: a width of the current video block, or a height of the current video block,

wherein the dimension of the template or the dimension of the partial of the template comprises at least one of: a width of the template or a width of the partial of the template, or a height of the template or a height of the partial of the template, or

wherein a location of the current video block or a location of the template comprises at least one of: a location of a center sample of the current video block, a location of a center sample of the template, a location of a top-left sample of the current video block, or a location of a top-left sample of the template.

4 . The method of claim 1 , wherein a current template of the current video block and the reference template of a reference block of the current video block are horizontal templates, a width of the horizontal templates being the same with a width of the current video block, and the sample reordering mode comprises a horizontal flip reordering mode,

wherein the current template comprises a neighboring sample above to the current video block, and the reference template comprises a neighboring sample above to the reference block,

wherein a location difference between the current template and the current video block is the same with a location difference between the reference template and the reference block,

wherein a first horizontal distance between a top-left sample of the current template and a top-left sample of the current video block is the same with a second horizontal distance between a top-left sample of the reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the current template and the top-left sample of the current video block is the same with a second vertical distance between the top-left sample of the reference template and the top-left sample of the reference block,

wherein the first and second horizontal distances are zero, and the first and second vertical distances are a height of the current template or the reference template, or

wherein a distance between the top-left sample of the current template and the top-left sample of the reference template is the same with a distance between the top-left sample of the current video block and the top-left sample of the reference block.

5 . The method of claim 4 , wherein at least one of a sample in the current template or a sample in the reference template is flipped,

wherein a flip type of the current video block is horizontal flip, and at least one of a sample in the current template or a sample in the reference template is flipped, or

wherein a horizontal distance between the current video block and the reference block is the same with a horizontal distance between the current template and the reference template.

6 . The method of claim 1 , wherein a current template of the current video block and the reference template of a reference block of the current video block are vertical templates, a height of the vertical templates being the same with a height of the current video block, and the sample reordering mode comprises a horizontal flip reordering mode,

wherein the current template comprises a neighboring sample left to the current video block, and the reference template comprises a neighboring sample right to the reference block,

wherein a location difference between the current template and the current video block is different from a location difference between the reference template and the reference block,

wherein a first horizontal distance between a top-left sample of the current template and a top-left sample of the current video block is different from a second horizontal distance between a top-left sample of the reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the current template and the top-left sample of the current video block is the same with a second vertical distance between the top-left sample of the reference template and the top-left sample of the reference block,

wherein the first horizontal distance is a width of the current template, and the first vertical distance is zero,

wherein the second horizontal distance is a width of the current video block, the reference template being right to the reference block, and the second vertical distance is zero, or

wherein a sum of a first distance between the top-left sample of the current template and the top-left sample of the reference template, a width of the current video block and a width of the current template is equal to a second distance between the top-left sample of the current video block and the top-left sample of the reference block.

7 . The method of claim 6 , wherein at least one of a sample in the current template or a sample in the reference template is flipped,

wherein a flip type of the current video block is horizontal flip, and at least one of a sample in the current template or a sample in the reference template is flipped, or

wherein a sum of a first horizontal distance between the current video block and the reference block, a width of the current video block and a width of the current template is equal to a second horizontal distance between the current template and the reference template.

8 . The method of claim 1 , wherein a current template of the current video block comprises a horizontal current template and a vertical current template, the reference template of a reference block of the current video block comprises a horizontal reference template and a vertical reference template, and the sample reordering mode comprises a horizontal flip reordering mode.

9 . The method of claim 8 , wherein a width of the horizontal current template and a width of the horizontal reference template is the same with a width of the current video block,

wherein the current template comprises a neighboring sample above to the current video block and a neighboring sample left to the current video block, and the reference template comprises a neighboring sample above to the reference block and a neighboring sample right to the reference block,

wherein a location difference between the horizontal current template and the current video block is the same with a location difference between the horizontal reference template and the reference block,

wherein a first horizontal distance between a top-left sample of the horizontal current template and a top-left sample of the current video block is the same with a second horizontal distance between a top-left sample of the horizontal reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the horizontal current template and the top-left sample of the current video block is the same with a second vertical distance between the top-left sample of the horizontal reference template and the top-left sample of the reference block,

wherein the first horizontal distance is zero, and the first vertical distance is a height of the horizontal current template,

wherein a distance between the top-left sample of the horizontal current template and the top-left sample of the horizontal reference template is equal to a distance between the top-left sample of the current video block and the top-left sample of the reference block,

wherein a location difference between the vertical current template and the current video block is different from a location difference between the vertical reference template and the reference block,

wherein a third horizontal distance between a top-left sample of the vertical current template and a top-left sample of the current video block is different from a fourth horizontal distance between a top-left sample of the vertical reference template and a top-left sample of the reference block, and a third vertical distance between the top-left sample of the vertical current template and the top-left sample of the current video block is the same with a fourth vertical distance between the top-left sample of the vertical reference template and the top-left sample of the reference block,

wherein the third horizontal distance is a width of the vertical current template, and the third vertical distance is zero,

wherein the fourth horizontal distance is a width of the current video block, the vertical reference template being right to the reference block, and the fourth vertical distance is zero,

wherein at least one of a sample in the horizontal current template or a sample in the horizontal reference template is flipped,

wherein at least one of a sample in the vertical current template or a sample in the vertical reference template is flipped,

wherein a flip type of the current video block is horizontal flip, and a sample in the horizontal current template and a sample in the vertical current template are flipped,

wherein a flip type of the current video block is horizontal flip, and a sample in the horizontal reference template and a sample in the vertical reference template are flipped, or

wherein a horizontal distance between the current video block and the reference block is equal to a horizontal distance between the horizontal current template and the horizontal reference template.

10 . The method of claim 8 , wherein a width of the horizontal current template and a width of the horizontal reference template are the same, a width of the vertical current template and a width of the vertical reference template are the same, and the width of the horizontal current template is equal to a sum of a width of the current video block and a width of the vertical current template,

wherein a location difference between the horizontal current template and the current video block is different from a location difference between the horizontal reference template and the reference block,

wherein a first horizontal distance between a top-left sample of the horizontal current template and a top-left sample of the current video block is different from a second horizontal distance between a top-left sample of the horizontal reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the horizontal current template and the top-left sample of the current video block is the same with a second vertical distance between the top-left sample of the horizontal reference template and the top-left sample of the reference block,

wherein the first horizontal distance is a width of the vertical current template, and the first vertical distance is a height of the horizontal current template,

wherein the second horizontal distance is zero, and the second vertical distance is a height of the horizontal current template,

wherein a distance between the top-left sample of the horizontal current template and the top-left sample of the horizontal reference template is equal to a sum of a width of the vertical current template and a distance between the top-left sample of the current video block and the top-left sample of the reference block,

wherein the third horizontal distance is a width of the vertical current template, and the third vertical distance is zero,

wherein the fourth horizontal distance is a width of the current video block, the vertical reference template being right to the reference block, and the fourth vertical distance is zero,

wherein at least one of a sample in the horizontal current template or a sample in the horizontal reference template is flipped,

wherein at least one of a sample in the vertical current template or a sample in the vertical reference template is flipped,

wherein a flip type of the current video block is horizontal flip, and a sample in the horizontal current template and a sample in the vertical current template are flipped,

wherein a flip type of the current video block is horizontal flip, and a sample in the horizontal reference template and a sample in the vertical reference template are flipped, or

wherein a horizontal distance between the horizontal current template and the horizontal reference template is equal to a sum of a width of the vertical current template and a horizontal distance between the current video block and the reference block.

11 . The method of claim 1 , wherein a current template of the current video block and the reference template of a reference block of the current video block are horizontal templates, a width of the horizontal templates being the same with a width of the current video block, and the sample reordering mode comprises a vertical flip reordering mode,

wherein the current template comprises a neighboring sample above to the current video block, and the reference template comprises a neighboring sample below to the reference block,

wherein a location difference between the current template and the current video block is different from a location difference between the reference template and the reference block,

wherein a first horizontal distance between a top-left sample of the current template and a top-left sample of the current video block is the same a second horizontal distance between a top-left sample of the reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the current template and the top-left sample of the current video block is different from a second vertical distance between the top-left sample of the reference template and the top-left sample of the reference block,

wherein the first horizontal distance is zero, and the first vertical distance is a height of the current template,

wherein the second horizontal distance is zero, and the second vertical distance is a height of the current video block, the reference template being below to the reference block,

wherein a sum of a first distance between the top-left sample of the current template and the top-left sample of the reference template, a height of the current video block and a height of the current template is equal to a second distance between the top-left sample of the current video block and the top-left sample of the reference block,

wherein at least one of a sample in the current template or a sample in the reference template is flipped,

wherein a flip type of the current video block is vertical flip, and at least one of a sample in the current template or a sample in the reference template is flipped, or

wherein a sum of a first vertical distance between the current video block and the reference block, a height of the current video block and a height of the current template is equal to a second vertical distance between the current template and the reference template.

12 . The method of claim 1 , wherein a current template of the current video block and the reference template of a reference block of the current video block are vertical templates, a height of the vertical templates being the same with a height of the current video block, and the sample reordering mode comprises a vertical flip reordering mode,

wherein the current template comprises a neighboring sample left to the current video block, and the reference template comprises a neighboring sample left to the reference block,

wherein a location difference between the current template and the current video block is the same with a location difference between the reference template and the reference block,

wherein the first and second horizontal distances are a width of the current template or the reference template, and the first and second vertical distances are zero,

wherein at least one of a sample in the current template or a sample in the reference template is flipped,

wherein a flip type of the current video block is vertical flip, and at least one of a sample in the current template or a sample in the reference template is flipped, or

wherein a vertical distance between the current video block and the reference block is the same with a vertical distance between the current template and the reference template.

13 . The method of claim 1 , wherein a current template of the current video block comprises a horizontal current template and a vertical current template, the reference template of a reference block of the current video block comprises a horizontal reference template and a vertical reference template, and the sample reordering mode comprises a vertical flip reordering mode.

14 . The method of claim 13 , wherein a width of the horizontal current template and a width of the horizontal reference template is the same with a width of the current video block,

wherein the current template comprises a neighboring sample above to the current video block and a neighboring sample left to the current video block, and the reference template comprises a neighboring sample below to the reference block and a neighboring sample left to the reference block,

wherein a first horizontal distance between a top-left sample of the horizontal current template and a top-left sample of the current video block is the same with a second horizontal distance between a top-left sample of the horizontal reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the horizontal current template and the top-left sample of the current video block is different from a second vertical distance between the top-left sample of the horizontal reference template and the top-left sample of the reference block,

wherein the first horizontal distance is zero, and the first vertical distance is a height of the horizontal current template,

wherein the second horizontal distance is zero, and the second vertical distance is a height of the current video block, the horizontal reference template being below to the reference block,

wherein a distance between the top-left sample of the horizontal current template and the top-left sample of the horizontal reference template is equal to a sum of a height of the current video block, a height of the horizontal current template and a distance between the top-left sample of the current video block and the top-left sample of the reference block,

wherein a location difference between the vertical current template and the current video block is the same with a location difference between the vertical reference template and the reference block,

wherein a third horizontal distance between a top-left sample of the vertical current template and a top-left sample of the current video block is the same with a fourth horizontal distance between a top-left sample of the vertical reference template and a top-left sample of the reference block, and a third vertical distance between the top-left sample of the vertical current template and the top-left sample of the current video block is the same with a fourth vertical distance between the top-left sample of the vertical reference template and the top-left sample of the reference block,

wherein the third horizontal distance is a width of the vertical current template, and the third vertical distance is zero,

wherein at least one of a sample in the horizontal current template or a sample in the horizontal reference template is flipped,

wherein at least one of a sample in the vertical current template or a sample in the vertical reference template is flipped,

wherein a flip type of the current video block is vertical flip, and a sample in the horizontal current template and a sample in the vertical current template are flipped,

wherein a flip type of the current video block is vertical flip, and a sample in the horizontal reference template and a sample in the vertical reference template are flipped, or

wherein a vertical distance between the horizontal current template and the horizontal reference template is equal to a sum of a height of the current video block, a height of the horizontal current template and a vertical distance between the current video block and the reference block.

15 . The method of claim 13 , wherein a width of the horizontal current template and a width of the horizontal reference template are the same, a width of the vertical current template and a width of the vertical reference template are the same, and the width of the horizontal current template is equal to a sum of a width of the current video block and a width of the vertical current template,

wherein the current template comprises a neighboring sample above to the current video block and a neighboring sample left to the current video block, and the reference template comprises a neighboring sample below to the reference block and a neighboring sample left to the reference block,

wherein a first horizontal distance between a top-left sample of the horizontal current template and a top-left sample of the current video block is the same with a second horizontal distance between a top-left sample of the horizontal reference template and a top-left sample of the reference block, and a first vertical distance between the top-left sample of the horizontal current template and the top-left sample of the current video block is different from a second vertical distance between the top-left sample of the horizontal reference template and the top-left sample of the reference block,

wherein the first horizontal distance is a width of the vertical current template, and the first vertical distance is a height of the horizontal current template,

wherein the second horizontal distance is a width of the vertical current template, and the second vertical distance is a height of the current video block, the horizontal reference template being below the reference block,

wherein a location difference between the vertical current template and the current video block is the same with a location difference between the vertical reference template and the reference block,

wherein the third horizontal distance is a width of the vertical current template, and the third vertical distance is zero,

wherein at least one of a sample in the horizontal current template or a sample in the horizontal reference template is flipped,

wherein at least one of a sample in the vertical current template or a sample in the vertical reference template is flipped,

wherein a flip type of the current video block is vertical flip, and a sample in the horizontal current template and a sample in the vertical current template are flipped,

wherein a flip type of the current video block is vertical flip, and a sample in the horizontal reference template and a sample in the vertical reference template are flipped, or

16 . The method of claim 1 , wherein the conversion includes encoding the current video block into the bitstream.

17 . The method of claim 1 , wherein the conversion includes decoding the current video block from the bitstream.

18 . An apparatus for processing video data comprising a processor and a non-transitory memory with instructions thereon, wherein the instructions upon execution by the processor, cause the processor to:

determine, for a conversion between a current video block of a video and a bitstream of the video, a reference template of the current video block based on coding information of the current video block, the current video block being coded with a sample reordering mode; and

perform the conversion based on the reference template.

19 . A non-transitory computer-readable storage medium storing instructions that cause a processor to:

perform the conversion based on the reference template.

20 . A non-transitory computer-readable recording medium storing a bitstream of a video which is generated by a method performed by a video processing apparatus, wherein the method comprises:

determining a reference template of a current video block of the video based on coding information of the current video block, the current video block being coded with a sample reordering mode; and

generating the bitstream based on the reference template.