Adaptive Video Transform Kernels for Intra-Mode Coding Efficiency

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 in video signals.

Innovation Solution

The method involves determining a transform kernel set for video signal processing based on intra template matching and intra prediction modes, using a set of transform matrices including multiple transform sets, low frequency non-separable transforms, and non-separable primary transforms, to enhance coding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional video compression techniques are used, then basic spatial and temporal correlation is handled, but coding efficiency is insufficient

Engineering Contradiction:
Improvecoding efficiencyVSAvoidprocessing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies dynamics by making the transform kernel selection adaptive rather than fixed. The transform kernel is dynamically selected based on the intra prediction mode of the current block, allowing the processing to adapt to different block characteristics and prediction modes, thereby improving coding efficiency without requiring a completely complex system redesign

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of transform kernel selection based on intra prediction mode. Different transform kernels are applied depending on the prediction mode value, optimizing the transformation process for different types of blocks. This parameter-based adaptation improves coding efficiency by matching the transform characteristics to the prediction characteristics

Inventive Principle:
Principle #35Parameter changes

2Productivity

If transform kernel set is determined based on intra prediction mode, then coding efficiency increases, but processing complexity increases

Engineering Contradiction:
Improvecoding efficiencyVSAvoidtransform processing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The transform kernel is selected based on the intra prediction mode parameter. The patent establishes a mapping relationship between prediction modes and transform kernels, where different mode ranges correspond to different kernel selections. This parameter-based approach improves coding efficiency while keeping the complexity manageable through systematic classification

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments the transform processing by dividing it into different kernel sets based on intra prediction mode categories. By segmenting the transform kernel selection into mode-based groups, the system handles complexity in a structured way, processing different block types with appropriate specialized kernels rather than using a single universal approach

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4589945A1Video signal processing method and device therefor
Publication Date: 2025.07.23 WILUS INSTITUTE OF STANDARDS & TECHNOLOGY INC
  • EP4589945A1 patent drawingFigure 1~2
  • EP4589945A1 patent drawingFigure 3~4
  • EP4589945A1 patent drawingFigure 5

AI summary

This processor of a video signal decoding device may: determine a first prediction mode of a current block, generate a prediction block of the current block on the basis of the first prediction mode, generate a residual block of the current block on the basis of a transform matrix set which is determined on the basis of a second prediction mode, and reconstruct the current block on the basis of the prediction block and the residual block.