Image Coding Transform Index for MIP and LFNST Interference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high resolution and high quality images/videos, such as 4K and 8K Ultra High Definition, poses a challenge due to the higher amount of information required for transmission and storage, leading to increased costs and complexity.
Innovation Solution
The proposed method enhances image/video coding efficiency by using a transform index for blocks applying Matrix-based Intra Prediction (MIP), allowing for efficient transformation and coding of these blocks, while minimizing interference with low frequency non-separable transform (LFNST).
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If high resolution and high quality image/video (4K, 8K UHD) are transmitted and stored, then image quality and resolution are improved, but transmission and storage costs increase
Solution Approach 1:
The patent applies different transform types (DST, DCT, DST-II) based on the intra prediction mode parameters. By changing the transform parameter according to the prediction mode, the coding efficiency is improved without increasing the amount of transmitted data, thus resolving the contradiction between image quality and data quantity
Solution Approach 2:
The patent dynamically selects the transform type based on the intra prediction mode. The transform type is not fixed but adapts to the characteristics of the prediction mode, allowing optimal compression for different image characteristics while maintaining high quality
2Manufacturing precision
If transform index information is signaled for all blocks, then transformation accuracy is improved, but coding complexity increases
Solution Approach 1:
The patent applies transform index signaling selectively based on the intra prediction mode. Instead of applying the same transform to all blocks, different transform types are applied to different regions (blocks) depending on their prediction mode characteristics, improving accuracy where needed while reducing complexity where not necessary
Solution Approach 2:
The patent signals transform index information only for specific intra predicted blocks where it provides benefit, rather than for all blocks. This partial application of transform indexing maintains transformation accuracy for critical blocks while reducing overall coding complexity
3Productivity
If MIP and LFNST are both applied to a block, then coding efficiency is improved, but interference between transforms increases
Solution Approach 1:
The patent prevents transform interference by establishing a rule that LFNST is not applied to blocks where MIP is already used. This preliminary prevention avoids the harmful interaction between the two transforms while maintaining coding efficiency through appropriate transform selection for non-MIP blocks
4Adaptability or versatility
If more transform types are supported, then adaptation to different image characteristics is improved, but processing complexity increases
Solution Approach 1:
The patent uses parameter changes (transform type selection based on intra prediction mode) to adapt to different image characteristics. By linking transform type to prediction mode parameters, the system achieves high adaptability without requiring complex decision logic, thus managing processing complexity
Data Source
AI summary
An image decoding method according to the present document comprises the steps of: receiving image information including residual information for a current block; and generating residual samples of the current block on the basis of the residual information, wherein the residual information includes information on transform coefficients of the current block, and the residual samples are generated from the information on the transform coefficients on the basis of low-frequency non-separable transform (LFNST) index information.