Transform Skip Mode Harmonization in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently controlling residual coding for video sequences, particularly in harmonizing transform skip mode and multiple transform selection, which affects encoding and decoding processes.
Innovation Solution
A method and apparatus that utilize a processor to control residual coding by identifying appropriate transforms based on a multiple transform selection index, enabling transform skip mode for coded blocks and selecting between discrete cosine transform, discrete sine transform, Hadamard transform, and Haar transform for horizontal and vertical transforms, depending on the index, to perform residual coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If transform skip mode is used for residual coding, then coding efficiency is improved for certain block types, but harmonization with multiple transform selection becomes complex and difficult to control
Solution Approach 1:
The patent merges transform skip mode and multiple transform selection into a unified transform selection framework. The encoder and decoder use a single mts_idx parameter to select from multiple transform types including DCT-2, DST-7, and identity transform, eliminating the need for separate transform skip flags and enabling unified control of transform operations.
Solution Approach 2:
The unified transform selection mechanism provides multi-functionality by enabling the same selection framework to handle both traditional DCT-2 transforms and transform skip operations. The identity transform serves as a universal option that can replace DCT-2 when transform skip is desired, allowing a single mechanism to perform multiple transform functions.
2Adaptability or versatility
If separate control mechanisms are used for transform skip mode and multiple transform selection, then each mode can be optimized independently, but the overall control structure becomes complex and difficult to manage
Solution Approach 1:
The patent combines separate control mechanisms into a unified structure where a single mts_idx parameter controls transform selection. This unified control structure maintains adaptability by allowing independent optimization of different transform types while simplifying the overall control architecture through consolidation.
Solution Approach 2:
The unified control mechanism provides universal functionality by handling both transform skip and multiple transform selection through the same parameter and decision logic. This enables independent optimization of each transform type while maintaining a simple, manageable control structure.
3Device complexity
If unified transform selection is implemented to harmonize transform skip mode and multiple transform selection, then control complexity is reduced, but backward compatibility with existing video coding standards may be affected
Solution Approach 1:
The patent maintains backward compatibility by preserving DCT-2 as the default transform type, which is widely supported in existing video coding standards. The unified transform selection allows local optimization by selecting identity transform or DST-7 only when beneficial, while defaulting to DCT-2 for compatibility with legacy decoders and common video content.
Solution Approach 2:
The implementation uses parameter changes to maintain compatibility, where mts_idx values are mapped to transform types in a way that preserves existing behavior for common cases. By carefully selecting which transform types are available and how they are selected, the patent achieves unified control while maintaining interoperability with existing video coding infrastructure.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A method of controlling residual coding for decoding or encoding of a video sequence, is performed by at least one processor and includes, based on a multiple transform selection (MTS) index indicating that a transform skip mode is enabled for a coded block of the video sequence, identifying an identity transform as each of a horizontal transform and a vertical transform. The method further includes, based on the MTS index indicating that the transform skip mode is not enabled for the coded block, identifying one of a discrete cosine transform (DCT), a discrete sine transform (DST), a Hadamard transform and a Haar transform, as either one or both of the horizontal transform and the vertical transform. The method further includes performing the residual coding of the coded block, using the identified horizontal transform and the identified vertical transform.