CTU Level Rate Control via Perceptual Distortion Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing rate control algorithms for High Efficiency Video Coding (HEVC) focus on mean square error (MSE) optimization, which may not be optimal for perceptual quality, and do not fully consider the human visual system in coding tree unit (CTU) level bit allocation.
Innovation Solution
A perceptual rate-distortion model using a divisive normalization framework is established to characterize the relationship between local visual quality and coding bits, transformed into a global optimization problem solved with convex optimization algorithms for optimal CTU level coding bit allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If MSE-based rate control optimization is used, then coding efficiency is improved, but perceptual quality is not optimized
Solution Approach 1:
The patent transforms the rate control optimization from MSE-based parameters to perceptual distortion-based parameters. By changing the objective function from minimizing mean square error to minimizing perceptual distortion measured by HVS models, the system achieves better perceptual quality while maintaining coding efficiency through the established perceptual R-D model.
Solution Approach 2:
The patent applies different perceptual weighting factors to different regions and components within CTUs based on their visual importance. By analyzing local characteristics such as gradient magnitude, texture complexity, and human visual system sensitivity, the system allocates bits locally according to perceptual needs rather than uniform MSE criteria, thereby improving overall perceptual quality.
2Device complexity
If traditional RC algorithms are used, then bit allocation is simplified, but human visual system characteristics are not considered
Solution Approach 1:
The patent segments the video content into CTUs and further into sub-blocks, applying different perceptual analysis and bit allocation strategies to each segment. This segmentation allows the system to consider local HVS characteristics without requiring complex global optimization, thereby balancing algorithm complexity with perceptual adaptability.
Solution Approach 2:
The patent performs preliminary analysis of perceptual characteristics (gradient, texture, HVS sensitivity) before the actual rate control decision. By pre-computing perceptual weights and distortion estimates for each CTU/sub-block, the system prepares the necessary information in advance, making the subsequent bit allocation process more efficient and better adapted to perceptual requirements.
3Ease of manufacture
If uniform bit allocation is used, then implementation is simpler, but perceptual distortion is not minimized
Solution Approach 1:
The patent implements non-uniform bit allocation at the CTU and sub-block level by computing local perceptual distortion estimates and applying region-specific weighting factors. Areas with higher visual importance (e.g., high gradient regions, textured areas) receive more bits, while less important areas receive fewer bits, thereby minimizing overall perceptual distortion while maintaining reasonable implementation complexity through localized processing.
Data Source
AI summary
A method based on CTU level rate-distortion optimization for rate control in video coding which can effectively improve the perceptual rate-distortion performance and coding efficiency is provided. Firstly, a perceptual rate-distortion model is established using a divisive normalization framework, which characterizes the relationship between local visual quality and coding bits. Subsequently, the established perceptual rate-distortion model is applied to overall distortion optimization which is transformed into a global optimization problem and solved with convex optimization algorithms to obtain optimal CTU level coding bit allocation.