Video Block Chroma Scaling for More Efficient Motion Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video signal processing methods lack efficiency in coding, particularly in handling spatial and temporal correlations, necessitating improved techniques for encoding and decoding.

Innovation Solution

A method and apparatus for video signal processing that includes transformation, quantization, prediction, and entropy coding, utilizing intra and inter prediction units, adaptive filtering, and multi-hypothesis prediction to enhance coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional video compression methods are used, then the basic coding function is maintained, but the coding efficiency is insufficient for modern media transmission requirements

Engineering Contradiction:
Improvecoding efficiencyVSAvoidvideo signal processing performance
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent divides the video block into multiple sub-blocks and applies different motion vector difference (MVD) modeling techniques to each sub-block. This segmentation allows the encoder to capture local motion characteristics more accurately, improving coding efficiency while maintaining reliability through targeted optimization of motion compensation in each segment.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces higher-level merge with MVD (MMVD) enabling information that allows different sub-blocks to have different MVD modeling characteristics. By applying local quality principles, the system optimizes motion vector prediction for each sub-block according to its specific motion characteristics, thereby improving overall coding efficiency without compromising the reliability of motion compensation in any particular region.

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If motion vector precision is increased to improve block reconstruction accuracy, then the coding efficiency improves, but the data transmission requirements increase

Engineering Contradiction:
Improvemotion vector accuracyVSAvoiddata transmission volume
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent changes the parameter representation by using MVD (motion vector difference) instead of absolute motion vectors. This parameter change allows the system to transmit only the difference values rather than full motion vector data, reducing the quantity of data transmitted while maintaining high motion vector accuracy through precise difference encoding.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies MVD modeling selectively to sub-blocks that require it, rather than uniformly across all blocks. This partial action approach allows the system to increase motion vector accuracy where needed for better reconstruction while avoiding unnecessary data transmission overhead in regions where simple motion compensation suffices.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4542988B1Video signal processing method and device
Publication Date: 2026.05.13 WILUS INSTITUTE OF STANDARDS & TECHNOLOGY INC
  • EP4542988B1 patent drawingFigure 1
  • EP4542988B1 patent drawingFigure 2
  • EP4542988B1 patent drawingFigure 3

AI summary

A method for decoding a video signal, comprising: obtaining chroma component format information from a higher-level bitstream, the chroma component format information indicating a chroma format of a chroma component of a current block; obtaining, based on the chroma component format information, width-related information (SubWidthC) which indicates a width of a chroma array in comparison with a width of a luma array of the current block and height-related information (SubHeightC) which indicates a height of the chroma array in comparison with a height of the luma array of the current block; obtaining x-axis scale information based on the width-related information or information on a color component of the current block; obtaining y-axis scale information based on the height-related information or information on the color component of the current block; and determining a weighting value based on code information of a left block and code information of an upper block.