Multiple Transform Kernels for IBC Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video signal processing methods lack efficiency in coding and signaling, particularly in determining transform kernels for intra block copy (IBC) modes and transform units based on luma maximum transform sizes.
Innovation Solution
A method and apparatus for efficiently determining transform kernels by checking preset conditions on prediction modes and coding unit sizes, parsing syntax elements, and applying discrete cosine transform type-2 (DCT-2) when necessary, with options for multiple transform selection (MTS) modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple transform kernels are used for different prediction modes, then coding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent applies different transform kernels (DST-VII, DCT-VIII, DCT-II) to different prediction modes (intra prediction, inter prediction, IBC mode) based on the local characteristics of each mode. This allows optimal transform selection for each specific prediction scenario, improving coding efficiency while managing complexity through targeted differentiation rather than universal complexity.
Solution Approach 2:
The patent changes the transform kernel parameter based on the prediction mode being used. By selecting from multiple transform kernels (DST-VII for intra mode, DCT-VIII for inter mode, DCT-II for IBC mode), the system adapts the transform parameter to match the characteristics of each prediction mode, resolving the contradiction between efficiency and complexity.
2Manufacturing precision
If transform kernel information is explicitly signaled for each block, then transform accuracy is improved, but signaling overhead increases
Solution Approach 1:
The patent performs preliminary action by establishing default transform kernel associations with prediction modes before actual coding. The transform kernel is determined based on the prediction mode that is already selected, eliminating the need for separate explicit signaling. This preliminary association reduces signaling overhead while maintaining transform accuracy.
Solution Approach 2:
The system uses the prediction mode information that is already being signaled for other purposes to automatically determine the transform kernel. Instead of requiring separate transform kernel signaling, the existing prediction mode data serves dual purposes, reducing redundant signaling overhead while maintaining accurate transform selection.
3Device complexity
If a single transform kernel is used for all modes, then device complexity is reduced, but coding efficiency deteriorates
Solution Approach 1:
The patent introduces dynamics by making the transform kernel selection adaptive to the prediction mode rather than static. The transform kernel changes dynamically based on which prediction mode is active (intra, inter, or IBC), allowing the system to maintain simplicity in implementation while achieving high compression efficiency through mode-dependent adaptability.
Data Source
AI summary
A video signal processing method may comprise the steps of: confirming a prediction mode applied to a current coding unit; confirming whether a plurality of preset conditions are satisfied on the basis of at least one of the prediction mode of the current coding unit and a size of the current coding unit; parsing a first syntax element indicating a transform kernel applied to a transform unit included in the current coding unit when the plurality of preset conditions are satisfied; determining a transform kernel applied to horizontal and vertical directions of the current transform unit on the basis of the first syntax element; and generating a residual signal of the current transform unit by performing an inverse transform on the current transform unit using the determined transform kernel.