Video Transform Skip Mode Flag and Matrix Index Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image encoding/decoding methods face limitations in achieving high objective and subjective quality due to the use of a single transform/inverse transform type or excessive signaling for various transform types, leading to increased data volume and transmission/storage costs.
Innovation Solution
A video encoding/decoding method that determines whether to apply a reduced secondary transform/inverse transform based on the intra-prediction mode, prediction mode, color component, size, and form of the current block, using a transform skip mode flag and transform matrix index to optimize transform operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single transform/inverse transform type is used, then device complexity is reduced, but image quality (objective and subjective) deteriorates
Solution Approach 1:
The patent implements dynamic transform selection by introducing a transform skip mode flag and transform matrix index that allow the transform type to be adaptively changed based on prediction mode and block characteristics. The system can switch between skip mode, primary transform, and secondary transform dynamically, resolving the contradiction between fixed complexity and variable quality requirements.
Solution Approach 2:
The patent changes the parameters of the transform operation by introducing different transform matrices (identity matrix for skip mode, primary transform matrices, and secondary transform matrices) and selecting among them based on prediction mode and block size. This parameter-based approach allows quality improvement without fundamentally changing the transform framework complexity.
2Manufacturing precision
If various transform/inverse transform types are used to improve image quality, then image quality improves, but signaling overhead increases
Solution Approach 1:
The patent applies transform skip mode and secondary transform selectively based on local characteristics such as prediction mode and block size. By using the transform skip mode flag and transform matrix index only where needed (intra-block copy mode, specific prediction modes, certain block sizes), the signaling overhead is minimized while quality is improved locally where beneficial.
Solution Approach 2:
The patent implements partial transform application by allowing skip mode (no transform) for certain blocks while applying primary or secondary transform for others. The transform skip mode flag and transform matrix index enable selective application, avoiding excessive signaling for blocks where transform would not provide benefit.
3Quantity of substance
If transform skip mode is applied, then data volume is reduced, but image quality may deteriorate
Solution Approach 1:
The patent dynamically selects between skip mode and transform based on the transform skip mode flag and transform matrix index. When skip mode is beneficial for compression, it is applied; when quality is prioritized, primary or secondary transform is applied. This dynamic selection resolves the contradiction between data volume reduction and quality maintenance.
Solution Approach 2:
The patent applies transform operations partially rather than universally. By using the transform skip mode flag to determine skip mode application and the transform matrix index to select among different transform types, the system applies transform only where needed, achieving compression where possible while maintaining quality where necessary.
Data Source
AI summary
Disclosed herein is a video decoding method including: obtaining a transform skip mode flag indicating whether transform/inverse transform is skipped in a current block; when transform/inverse transform is skipped in the current block according to the transform skip mode flag, determining that secondary transform/inverse transform is skipped in the current block; and when transform/inverse transform is not skipped in the current block according to the transform skip mode flag, obtaining a transform matrix index for secondary transform/inverse transform of the current block and, determining whether secondary transform/inverse transform is skipped in the current block based on the transform matrix index.