Hierarchical Syntax Control for Video Filtering Adaptability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently processing and decoding video data due to increased bandwidth demands and complex picture partitioning schemes.
Innovation Solution
The proposed solution involves techniques for video encoders and decoders to process coded video representations using control information. This includes methods for converting video pictures and bitstreams based on specific rules, such as syntax element indications for coding tools like Luma Mapping with Chroma Scaling (LMCS) and Sample Adaptive Offset (SAO), and determining the usage of scaling tools based on the presence of single slices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If complex picture partitioning schemes are used to improve video processing flexibility, then adaptability is improved, but device complexity increases
Solution Approach 1:
The video picture is divided into multiple slices, which are further divided into picture regions. This hierarchical segmentation allows independent processing and filtering control for different regions, improving adaptability while managing complexity through modular structure.
Solution Approach 2:
The filtering tool usage is dynamically controlled through syntax flags that can be set at different levels (picture level, slice level, region level). This dynamic control allows the system to adapt filtering application to specific content requirements without hardcoding complex decision logic.
2Adaptability or versatility
If multiple syntax flags are used to control filtering tools at different levels, then adaptability is improved, but device complexity increases
Solution Approach 1:
The control structure is extended to multiple hierarchical levels (picture level, slice level, region level), adding a dimensional aspect to filtering control. This allows coarser control at higher levels and finer control at lower levels, managing complexity through hierarchical abstraction.
Solution Approach 2:
Different filtering tools and parameters can be applied to different picture regions based on local content characteristics. Syntax flags enable region-specific control of filtering tools, allowing adaptive processing where different areas of the picture receive different treatment based on their specific requirements.
3Manufacturing precision
If filtering tools are applied to all picture regions, then manufacturing precision is improved, but loss of energy increases
Solution Approach 1:
Instead of applying filtering tools uniformly to all picture regions, the system applies filtering selectively to specific regions where it is most beneficial. Syntax flags control which regions receive filtering processing, avoiding unnecessary computation in regions where filtering would provide minimal benefit.
Solution Approach 2:
Filtering tools are applied with different intensities and types to different picture regions based on local content characteristics. Regions with high detail or importance receive more aggressive filtering, while regions with simple content receive minimal or no filtering, optimizing the balance between quality and energy consumption.
Data Source
AI summary
Several techniques for video encoding and video decoding are described. One example method includes performing a conversion between a video picture of a video comprising one or more slices and a bitstream of the video picture according to a rule. The rule specifies that that whether a first syntax element indicating a usage of a coding tool is present at a first level is based on a syntax flag indicating whether a syntax structure of a second level is not present at the first level, wherein the second level is higher than the first level, and wherein the second level is the video picture level or higher than the video picture level.