VVC Picture Header Syntax for Lower Bitrate Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The introduction of a mandatory picture header in the Versatile Video Coding (VVC) standard increases bitrate, particularly in low bitrate instances, due to unnecessary signaling of syntax elements when most pictures contain only one slice type.
Innovation Solution
Modify the picture header signaling by merging delta QP signaling for Inter and Intra slices, using two override flags for partitioning parameters, and adding an override flag for motion information, allowing the decoder to skip unnecessary syntax elements when all slices in a picture use the same coding mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a mandatory picture header is introduced to provide flexibility in decoding slices, then adaptability is improved, but bitrate increases due to unnecessary signaling of syntax elements
Solution Approach 1:
The patent extracts only the necessary syntax elements for the actual slice type present in the picture. When all slices are of the same type (e.g., all intra slices or all inter slices), the picture header signals only the relevant syntax elements for that type, removing unnecessary syntax elements for other slice types. This reduces bitrate while maintaining the adaptability to handle different slice types when needed.
Solution Approach 2:
The picture header structure is made dynamic by allowing it to adapt its content based on the actual slice configuration. The header can selectively include or exclude syntax elements depending on whether the picture contains mixed slice types or uniform slice types, optimizing the bitrate for each case while preserving flexibility.
2Adaptability or versatility
If separate syntax elements are signaled for Inter and Intra slices in the picture header, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent removes the complexity of handling separate syntax elements for both Inter and Intra slices by extracting only the necessary set based on the actual slice type. The decoder no longer needs to process and conditionally apply multiple sets of syntax elements, reducing computational complexity while maintaining support for multiple slice types through selective signaling.
3Adaptability or versatility
If all syntax elements are signaled in the picture header to ensure full flexibility, then adaptability is improved, but loss of information is reduced (redundancy increased)
Solution Approach 1:
The patent extracts and signals only the essential syntax elements required for the actual slice type in the picture. By removing redundant syntax elements for slice types that are not present, the patent eliminates unnecessary bitrate consumption while maintaining complete flexibility to signal all necessary elements when mixed slice types are used.
Data Source
AI summary
A method of decoding video data from a bitstream, the bitstream comprising video data corresponding to multiple slices, wherein the video bitstream includes a picture header; the method comprising: determining a coding mode for at least one slice; determining a set of syntax elements to be used for said coding mode from said picture header; and decoding said at least one slice using said determined syntax elements.