Image Decoding Using Block-Size Quantization Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing HEVC and VVC encoding methods lack effective quantization control for zeroed-out orthogonal transform coefficients, leading to suboptimal subjective image quality due to the inability to adjust quantization based on frequency components.
Innovation Solution
An image encoding and decoding apparatus that uses different quantization matrices for different block sizes within a single image, allowing for separate quantization and inverse quantization of orthogonal transform coefficients, thereby enabling improved subjective image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If zeroing out technique is used to reduce code amount, then encoding efficiency is improved, but quantization control based on frequency components is lost
Solution Approach 1:
The patent applies local quality by using different quantization matrices for different block sizes (e.g., 8x8, 16x16, 32x32, 64x64). Each block size has its own customized quantization matrix that is optimized for that specific size, allowing frequency-component-based quantization control to be maintained even when zeroing out is applied. This resolves the contradiction by preserving adaptability through localized customization while maintaining the efficiency gains of zeroing out.
Solution Approach 2:
The patent segments the quantization process by dividing the image into different block sizes and applying separate quantization matrices to each segment. This segmentation allows the zeroing out technique to be applied at the block level while maintaining quantization control through the use of multiple specialized quantization matrices, thus resolving the contradiction between encoding efficiency and quantization control capability.
2Device complexity
If a single quantization matrix is used for all block sizes, then device complexity is reduced, but subjective image quality deteriorates
Solution Approach 1:
The patent implements local quality by creating separate quantization matrices for different block sizes (8x8, 16x16, 32x32, 64x64), where each matrix is optimized for its specific block size. This allows the system to maintain high subjective image quality through customized quantization for each block size while managing complexity through a systematic approach to matrix generation and selection.
Solution Approach 2:
The patent changes the parameter of quantization matrix size and characteristics to match the block size being processed. By adjusting the quantization matrix parameters (size, weighting factors) according to the block size, the system achieves optimal image quality without excessive complexity, as the matrices follow predictable patterns based on frequency component characteristics.
3Speed
If quantization is performed without block size consideration, then processing speed is improved, but encoding efficiency deteriorates
Solution Approach 1:
The patent segments the quantization process by block size, with each block size having its own dedicated quantization matrix. This segmentation allows the system to maintain fast processing speeds through direct matrix application while improving encoding efficiency by optimizing each matrix for its specific block size, thus resolving the contradiction between speed and efficiency.
Solution Approach 2:
The patent performs preliminary action by pre-calculating and storing multiple quantization matrices for different block sizes before the actual encoding process. This allows the encoding system to quickly select and apply the appropriate matrix without complex calculations during processing, thus maintaining high processing speed while achieving optimal encoding efficiency through pre-optimized matrices.
Data Source
AI summary
Decoding corresponding to a first array of quantized coefficients including an N×M array corresponding to a first block and data corresponding to a second array including an N×M array corresponding to a second block. Deriving a first array of orthogonal transform coefficients from the first array of quantized coefficients by using at least a first quantization matrix of an N×M array of elements, and derives a second array of orthogonal transform coefficients from the second array of by using at least a second quantization matrix of an N×M array of elements. Performing inverse orthogonal transform on the first array of orthogonal transform coefficients to generate a P×Q array of pixels of first prediction residuals, and performs inverse orthogonal transform on the second array of orthogonal transform coefficients to generate an N×M array of pixels of second prediction residuals.