Independent Transform Kernel Selection for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video signal processing methods are inefficient in encoding and decoding due to limitations in transform kernel selection, which affects coding efficiency.
Innovation Solution
A method and apparatus for video signal processing that determines horizontal and vertical transform kernels based on the length of the sides of a transform block, allowing for independent selection of kernels regardless of the other side's length, and uses these kernels for inverse transformation to improve coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a single transform kernel is used for all transform blocks, then device complexity is reduced, but coding efficiency deteriorates
Solution Approach 1:
The patent applies local quality by selecting different transform kernels (e.g., DCT, DST, DRT) based on the specific characteristics of each transform block, particularly the length of its sides. This allows each block to undergo transformation with a kernel optimally suited to its local properties, thereby improving coding efficiency without requiring complex global adaptation mechanisms.
Solution Approach 2:
The patent changes the parameter of transform kernel type based on the side length of transform blocks. By categorizing blocks according to their dimensions and assigning appropriate kernels (such as using DST for blocks with specific side length characteristics), the system adapts the transformation process to block-specific parameters, enhancing efficiency while maintaining manageable complexity through rule-based selection.
2Adaptability or versatility
If transform kernel selection is based on both horizontal and vertical side lengths, then coding efficiency is improved, but device complexity increases
Solution Approach 1:
The patent segments the kernel selection process by independently determining horizontal and vertical transform kernels based on the respective side lengths of transform blocks. This segmentation allows each dimension to be processed separately with its own kernel selection rules, improving adaptability while reducing overall complexity compared to joint two-dimensional kernel selection.
Solution Approach 2:
The patent introduces asymmetry by allowing different transform kernels to be selected for horizontal and vertical transformations based on different side length criteria. This asymmetric approach enables optimized kernel selection for each dimension independently, capturing directional characteristics of video content more effectively while maintaining simpler determination rules for each axis.
Data Source
AI summary
Disclosed is a video signal processing method comprising the steps of: obtaining at least one transform block for a residual signal of a current block from a video signal bitstream; determining, on the basis of length information of a first side of the transform block, a horizontal transform kernel for horizontal transformation of the transform block, regardless of the length of a second side of the transform block, which is orthogonal to the first side; determining, on the basis of length information of the second side, a vertical transform kernel for vertical transformation of the transform block, regardless of the length of the first side; obtaining the residual signal of the current block by performing, on the transform block, inverse transformation using the horizontal transform kernel and the vertical transform kernel; and reconstructing the current block on the basis of the obtained residual signal.