Image Codec Quantization for Zero-Residual Nonzero Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding protocols generate all-zero residual coefficient blocks only when all quantized coefficients are zero, and non-zero residual coefficient blocks only when at least one quantized coefficient is non-zero, limiting flexibility and efficiency in encoding and decoding processes.
Innovation Solution
Adjust quantized coefficients and parameters in blocks with all-zero coefficients to generate non-zero quantized blocks, allowing all-zero residual blocks upon decoding, using base coefficient matrices and dequantization formulas to ensure all residual coefficients become zero.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If quantized coefficients are adjusted to non-zero values, then encoding flexibility is improved, but decoding complexity increases
Solution Approach 1:
The encoder pre-calculates and stores base coefficient matrices before actual encoding. These pre-computed matrices are used during decoding to efficiently recover original coefficients from adjusted quantized coefficients, reducing decoding complexity while maintaining encoding flexibility
Solution Approach 2:
Base coefficient matrices serve as intermediary structures that bridge the adjusted quantized coefficients and original transform coefficients. The matrices enable the decoder to reconstruct original coefficients through simple matrix operations without requiring complex inverse transformations
2Productivity
If quantized coefficient blocks are adjusted to include non-zero coefficients, then data compression efficiency is improved, but processing complexity increases
Solution Approach 1:
Instead of adjusting all quantized coefficients uniformly, the method selectively modifies only specific coefficients based on local characteristics of the image block. This localized adjustment reduces processing complexity while maintaining compression efficiency
Solution Approach 2:
The method changes the parameter representation by introducing base coefficient matrices that transform the relationship between quantized coefficients and original coefficients. This parameter transformation simplifies the adjustment process and reduces computational complexity
Data Source
AI summary
An image processing method and a codec are disclosed. The method includes: determining a quantized coefficient block obtained by quantizing a transform unit of an image; if quantized coefficients in the quantized coefficient block are all zeros, determining a quantized parameter threshold of the quantized coefficient block based on a base coefficient matrix for transforming the transform unit; and adjusting quantized coefficients at M locations in the quantized coefficient block based on the quantized parameter threshold. None of the adjusted quantized coefficients at the M locations is zero, and coefficients in a reconstructed residual coefficient block obtained by dequantizing and inversely transforming an adjusted quantized coefficient block are all zeros. According to the method, when the quantized coefficients are all zeros, a quantized coefficient block including a non-zero coefficient can be generated, and reconstructed residual coefficients obtained by dequantizing and inversely transforming the quantized coefficient block including the non-zero coefficient are all zeros.