Image Residual Quantization by Channel and Spatial Importance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image encoding and decoding technologies suffer from poor quantizer performance due to the lack of consideration for differences in feature channels and image textures during quantization, leading to suboptimal compression ratios and signal loss.
Innovation Solution
The method involves obtaining feature channel and spatial point differentiation information to construct dequantization and quantization precision parameters, allowing for tailored quantization and dequantization processes based on these differences, thereby improving encoding and decoding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If quantization is applied to residuals without considering feature channel and spatial point differences, then compression ratio is improved, but quantizer performance deteriorates
Solution Approach 1:
The patent applies different quantization precision parameters to different feature channels and spatial points based on their importance. Specifically, it constructs dequantization precision parameters corresponding to different feature channels and spatial points, allowing important features to retain higher precision while less important features use lower precision, thus resolving the contradiction between compression ratio and quantizer performance
Solution Approach 2:
The patent changes the quantization parameter from a single uniform value to multiple differentiated values based on feature channel and spatial point characteristics. By constructing dequantization precision parameters that vary across different features and locations, the system optimizes both compression efficiency and reconstruction quality simultaneously
2Device complexity
If uniform quantization precision is used for all features, then device complexity is reduced, but feature reconstruction accuracy deteriorates
Solution Approach 1:
The patent implements local quality by assigning different dequantization precision parameters to different feature channels and spatial points. This allows the system to maintain high reconstruction accuracy for important features while using lower precision for less critical ones, avoiding the need for uniformly high precision across all features
Data Source
AI summary
The present disclosure relates to the technical field of image processing and discloses an image decoding and encoding method and apparatus, a device and a storage medium. In the present disclosure, feature channel distinguishing information and/or spatial point distinguishing information are obtained, and image code streams are decoded to obtain quantized residual values; an inverse quantization precision parameter value is set for each quantized residual value according to the feature channel distinguishing information and/or the spatial point distinguishing information; inverse quantization is performed on the quantization residual values according to the inverse quantization precision parameter values to obtain reconstructed residual values; synthetic transformation is performed on the reconstructed residual values to obtain a reconstructed image block.