Quantization Matrix Derivation Using Shared Base Scaling for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards, such as HEVC, face challenges in efficiently representing and adapting quantization matrices for various transform block sizes and modes, particularly in the emerging VVC standard, which requires more flexible and efficient quantization matrix representation to handle increased block sizes and partition shapes.
Innovation Solution
The proposed solution involves deriving and representing quantization matrices using a shared base scaling matrix, with methods such as up-sampling and interpolation to generate matrices for larger sizes, and allowing for user-defined matrices with lossless entropy coding, to support adaptive multiple core transforms and various block sizes, enabling efficient frequency-dependent scaling and reduced memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If separate quantization matrices are used for different block sizes and modes, then coding precision is improved, but device complexity and memory requirements increase
Solution Approach 1:
The quantization matrix is segmented into a base matrix and scaling factors. The base matrix contains the fundamental quantization characteristics, while scaling factors adjust the matrix for different block sizes and modes. This segmentation allows the system to maintain multiple quantization configurations without storing complete separate matrices for each case.
Solution Approach 2:
A single base quantization matrix is designed to serve multiple functions across different block sizes and coding modes. By combining this universal base matrix with mode-specific scaling factors, the system achieves adaptive quantization for various scenarios (e.g., 4×4, 8×8, 16×16 blocks and different intra/inter modes) without requiring separate dedicated matrices for each case.
2Adaptability or versatility
If multiple quantization matrices are stored for different block sizes, then adaptability is improved, but memory requirements increase
Solution Approach 1:
The quantization matrix structure is organized in a nested manner where a base matrix is defined first, and then scaling factors are applied to derive matrices for larger block sizes. For example, an 8×8 base matrix serves as the foundation, and scaling factors are used to derive 16×16 and 32×32 matrices. This nested structure allows the system to maintain adaptability across different block sizes while storing only the base matrix and scaling factors in memory.
3Productivity
If frequency-dependent scaling is applied, then coding efficiency is improved, but computational complexity increases
Solution Approach 1:
Frequency-dependent scaling is applied selectively to different frequency components of the quantization matrix rather than uniformly across all frequencies. The scaling factors are designed to adjust specific frequency regions (e.g., low-frequency vs. high-frequency coefficients) based on their importance and characteristics. This local quality approach maintains coding efficiency by preserving important frequency information while applying appropriate scaling, without requiring complex computations across the entire frequency spectrum.
Data Source
AI summary
A method and apparatus for video coding using a coding mode belonging to a mode group comprising an Intra Block Copy (IBC) mode and an Intra mode are disclosed. According to the present invention, for both IBC and Intra mode, a same default scaling matrix is used to derive the scaling matrix for a current block. In another embodiment, for the current block with block size of M×N or N×M, and M greater than N, a target scaling matrix is derived from an M×M scaling matrix by down-sampling the M×M scaling matrix to an M×N or N×M scaling matrix.