Picture Header Syntax for Adaptive Video Coding Modes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression technologies, such as MPEG-2, MPEG-4, ITU-T.263, ITU-T.264/MPEG-4 AVC, ITU-T.265 HEVC, and VVC, require further improvements in coding efficiency.
Innovation Solution
A method and apparatus for video processing that includes syntax elements indicating residual and variance scale (RVS) mode, skip mode, and latent scale before synthesis (LSBS) mode in the picture header to simplify the synthesis transform module while maintaining reconstruction capability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If multiple syntax elements (RVS mode, skip mode, LSBS mode) are included in the picture header, then the synthesis transform module complexity is reduced, but the bitstream overhead increases
Solution Approach 1:
The patent merges multiple syntax elements (RVS mode flag, skip mode flag, and LSBS mode flag) into a single picture header structure. This consolidation allows the decoder to efficiently access all three mode indicators without processing separate, scattered syntax elements, thereby reducing the overall complexity of the synthesis transform module while maintaining comprehensive reconstruction capability.
Solution Approach 2:
The picture header is designed to serve multiple functions by incorporating syntax elements for different transform modes (RVS, skip, and LSBS) within a single structure. This multi-functional approach allows the header to control various reconstruction strategies simultaneously, reducing the need for separate control mechanisms and simplifying the overall decoding architecture.
2Device complexity
If the synthesis transform module is simplified, then the reconstruction capability is maintained, but the coding efficiency deteriorates
Solution Approach 1:
The patent implements dynamic mode selection by providing flags that enable or disable different transform modes (RVS, skip, LSBS) on a per-picture basis. This dynamic approach allows the encoder to adaptively choose the most efficient transform mode for each picture or sequence of pictures, maintaining high coding efficiency while keeping the synthesis transform module structure simple and manageable.
Solution Approach 2:
The patent changes the operational parameters of the synthesis transform module by introducing control flags that switch between different transform modes. By modifying these parameters (mode selection flags) rather than changing the fundamental module structure, the system maintains coding efficiency through adaptive parameter adjustment while preserving structural simplicity.
3Manufacturing precision
If multiple transform modes are supported, then the reconstruction quality is improved, but the computational complexity increases
Solution Approach 1:
The patent applies different transform modes to different regions or pictures based on their specific characteristics. By allowing selective application of RVS, skip, or LSBS modes through picture header flags, the system optimizes reconstruction quality for each local region or picture type without requiring all modes to be simultaneously active, thereby reducing overall computational complexity while maintaining high reconstruction quality where needed.
Data Source
AI summary
Embodiments of the disclosure provide a solution for video processing. A method for video processing is proposed. The method includes: performing a conversion between a video unit of a video and a bitstream of the video according to a rule, wherein the rule indicates that a first syntax element indicating whether a residual and variance scale (RVS) mode being enabled or not, a second syntax element indicating whether a skip mode being enabled or not, and a third syntax element indicating whether a latent scale before synthesis (LSBS) mode being enabled or not are included in a picture header.