Matrix-based Intra Prediction Selective Application for Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing demand for high-resolution, high-quality image/video data, such as 4K or 8K, poses challenges in efficient compression, transmission, and storage due to the high amount of information required, and existing matrix-based intra prediction (MIP) methods need improvements for enhanced efficiency and visual quality.
Innovation Solution
The proposed solution involves a method and apparatus for enhancing image/video coding efficiency by adaptively selecting the application of MIP based on the type of current block, reducing the number of MIP modes for blocks with low prediction efficiency, and simplifying signaling conditions for hardware implementation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If MIP process is applied to all block types, then prediction accuracy is improved, but computational complexity and signaling overhead increase
Solution Approach 1:
The patent applies MIP process selectively to specific block types (e.g., non-square blocks, blocks with specific size ratios) rather than uniformly to all blocks. This local quality approach ensures high prediction accuracy is achieved only where it provides benefit, while avoiding unnecessary computational complexity for block types where MIP would not improve results.
Solution Approach 2:
The patent introduces conditional parameters (block width, block height, size ratio thresholds) that dynamically control whether MIP is applied. By changing the application parameters of MIP based on block characteristics, the system achieves high prediction accuracy for suitable blocks while maintaining low complexity for unsuitable blocks.
2Measurement precision
If MIP process is applied to all block types, then prediction accuracy is improved, but transmission data increases
Solution Approach 1:
The patent applies MIP selectively to specific block types rather than all blocks, which reduces the amount of MIP-related signaling data that needs to be transmitted. By limiting MIP application to blocks where it provides benefit, the patent minimizes transmission overhead while maintaining prediction accuracy where needed.
Solution Approach 2:
The patent applies MIP process partially - only to blocks that meet specific criteria (non-square blocks, blocks with size ratio ≥ 2:1, etc.) rather than applying it excessively to all blocks. This partial action approach achieves sufficient prediction accuracy for difficult-to-compress blocks while avoiding unnecessary transmission data for blocks where simple prediction modes suffice.
3Ease of manufacture
If MIP is applied without block type limitations, then hardware implementation is simplified, but compression efficiency decreases
Solution Approach 1:
The patent uses simple parameter-based conditions (block width, block height, size ratio comparisons) to control MIP application. These parameter changes are easy to evaluate in hardware while still achieving good compression efficiency by targeting blocks where MIP provides the most benefit.
Solution Approach 2:
The patent segments the block processing into two paths: blocks that meet MIP criteria (non-square, specific size ratios) and blocks that don't. This segmentation allows hardware to efficiently route blocks through appropriate processing paths, maintaining implementation simplicity while improving overall compression efficiency through selective MIP application.
Data Source
AI summary
An image decoding method according to this disclosure includes receiving image information including intra matrix-based intra prediction (MIP) flag information related to whether an MIP is applied to a current block through a bitstream, deriving an intra prediction type for the current block as the MIP based on the intra MIP flag information, generating intra prediction samples for the current block based on the MIP, and generating reconstructed samples for the current block based on the intra prediction samples.