Systems and methods for inter prediction compensation

Patent metadata

Field	Value
Publication number	US-12598290-B2
Application number	US-202418441552-A
Country	US
Kind code	B2
Filing date	Feb 14, 2024
Priority date	Feb 21, 2021
Publication date	Apr 7, 2026
Grant date	Apr 7, 2026

Abstract

A video processing method includes: determining whether an inter predictor correction is enabled for a coding block; and when the inter predictor correction is enabled for the coding block, performing the inter predictor correction by: obtaining a plurality of predicted samples from a top boundary and a left boundary of a predicted block corresponding to the coding block; obtaining a plurality of reconstructed samples from top neighboring reconstructed samples and left neighboring reconstructed samples of the coding block; and deriving a corrected predicted block based on the plurality of the predicted samples, the plurality of the reconstructed samples and the predicted block.

First claim

What is claimed is:

1 . A video processing method, comprising:
determining whether an inter predictor correction is enabled for a coding block; and when the inter predictor correction is enabled for the coding block, performing the inter predictor correction by:
obtaining a plurality of predicted samples from a top boundary and a left boundary of a predicted block corresponding to the coding block, wherein the predicted samples are within the predicted block;
obtaining a plurality of reconstructed samples from top neighboring reconstructed samples and left neighboring reconstructed samples of the coding block; and
deriving a corrected predicted block based on the plurality of the predicted samples, the plurality of the reconstructed samples, and the predicted block,
wherein obtaining the plurality of the predicted samples and the plurality of the reconstructed samples comprises:
determining a first parameter based on a coding mode of the coding block;
determining a second parameter according to a width of the coding block and the first parameter;
determining a third parameter according to a height of the coding block and the first parameter; and
obtaining the plurality of the predicted samples and the plurality of the reconstructed samples based on the second parameter and the third parameter.

2 . The video processing method of claim 1 , wherein deriving the corrected predicted block further comprises:
deriving one or more parameters for the inter predictor correction based on the plurality of the predicted samples and the plurality of the reconstructed samples; and deriving the corrected predicted block based on the one or more parameters and the predicted block.

3 . The video processing method of claim 1 , wherein the first parameter comprises a horizontal position parameter and a vertical position parameter, and obtaining the plurality of the predicted samples and the plurality of the reconstructed samples further comprises:
determining a plurality of horizontal positions based on the horizontal position parameter or determining a plurality of vertical positions based on the vertical position parameter; obtaining the plurality of the predicted samples from the top boundary of the predicted block based on the plurality of the horizontal positions or obtaining the plurality of the predicted samples from the left boundary of the predicted block based on the plurality of the vertical positions; and obtaining the plurality of the reconstructed samples from the top neighboring reconstructed samples based on the plurality of the vertical positions or and obtaining the plurality of the reconstructed samples from the left neighboring reconstructed samples based on the plurality of the vertical positions.

4 . The video processing method of claim 3 , wherein a value of the horizontal position parameter is equal to a value of the vertical position parameter.

5 . The video processing method of claim 3 , wherein a value of the horizontal position parameter or a value of the vertical position parameter is equal to 4, 8 or 16.

6 . The video processing method of claim 5 , further comprising:
if the coding mode of the coding block indicates the predicted block is derived on sub-block level, setting the value of the horizontal position parameter and the value of the vertical position parameter to be equal to 8; otherwise, setting the value of the horizontal position parameter and the value of the vertical position parameter to be equal to 16; or if the coding mode of the coding block indicates the predicted block is derived on sub-block level, setting the value of the horizontal position parameter and the value of the vertical position parameter to be equal to 4; otherwise setting the value of the horizontal position parameter and the value of the vertical position parameter to be equal to 8.

7 . The video processing method of claim 3 , wherein the second parameter comprises a clipped value of the horizontal position parameter, and the third parameter comprises a clipped value of the vertical position parameter, the video processing method further comprising:
if a width of the coding block is greater than a value of the horizontal position parameter, setting the clipped value of the horizontal position parameter to be equal to the value of the horizontal position parameter; otherwise, setting the clipped value of the horizontal position parameter to be equal to the width of the coding block; if a height of the coding block is greater than a value of the vertical position parameter, setting the clipped value of the vertical position parameter to be equal to the value of the vertical position parameter; otherwise, setting the clipped value of the vertical position parameter to be equal to the height of the coding block.

8 . The video processing method of claim 1 , further comprising:
determining a first difference value based on the plurality of the predicted samples and a second difference value based on the plurality of the reconstructed samples; and deriving the corrected predicted block based on the first difference value and the second difference value.

9 . The video processing method of claim 8 , further comprising:
if the first difference value or the second difference value is less than a threshold, setting each corrected predicted sample in the corrected predicted block to be equal to a first value; otherwise deriving the corrected predicted block as the predicted block.

10 . The video processing method of claim 9 , wherein the threshold is dependent on a bit depth of a sample value.

11 . The video processing method of claim 10 , wherein the threshold is 1<<(Bitdepth-8), wherein “Bitdepth” is the bit depth of the sample value and “<<” denotes a left bit shift operation.

12 . The video processing method of claim 8 , further comprising:
sorting the plurality of the predicted samples and obtaining two smaller predicted samples and two greater predicted samples; deriving a first minimum value as an average of the two smaller predicted samples and a second minimum value as an average of two reconstructed samples respectively corresponding to the two smaller predicted samples; deriving a first maximum value as an average of the two greater predicted samples and a second maximum value as an average of two reconstructed samples respectively corresponding to the two greater predicted samples; and determining the first difference value as a difference between the first maximum value and the first minimum value, and determining the second difference value as a difference between the second maximum value and the second minimum value.

13 . The video processing method of claim 12 , further comprising:
if the first difference value or the second difference value is less than a threshold, setting each corrected predicted sample in the corrected predicted block to be equal to an averaged value of the second minimum value and the second maximum value; otherwise deriving the corrected predicted block as the predicted block.

14 . A method for signaling a bitstream, the method comprising:
receiving video data; encoding the video data to generate a bitstream by:
encoding an enabling flag associated with the video data, the enabling flag indicating whether an inter predictor correction is enabled for a coding block for an inter prediction process; and
encoding an index associated with the inter predictor correction based on which the inter predictor correction is performed;
wherein the enabled inter predictor correction is performed by:
obtaining a plurality of predicted samples within a predicted

Assignees

• Alibaba China Co Ltd

Inventors

• Li Xinwei
• Chen Jie
• Liao Ru-Ling
• Ye Yan

Classifications

• H04N19/132

  Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking · CPC title

• H04N19/159

  Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/593

  involving spatial prediction techniques · CPC title

• H04N19/51

  Motion estimation or motion compensation · CPC title