Video Decoder IntraTMP Fusion for Noisy Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital video compression technologies face challenges in improving decoding performance, especially in scenarios with complex pixel changes and noise, where IntraTMP modes may reduce decoding efficiency due to inaccurate prediction blocks.
Innovation Solution
A decoding method that combines intra template matching prediction (IntraTMP) with another prediction mode, such as intra or inter prediction, to fuse prediction blocks using weighted fusion, thereby improving prediction accuracy and decoding performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If IntraTMP mode is used for prediction, then decoding complexity is reduced, but prediction accuracy deteriorates in scenarios with complex pixel changes and noise
Solution Approach 1:
The patent implements dynamic prediction mode selection that adapts to different block characteristics. The decoder determines whether to use IntraTMP mode or alternative prediction modes based on template matching results and block-specific conditions, allowing the system to switch between low-complexity and high-accuracy modes dynamically rather than using a fixed approach
Solution Approach 2:
The patent changes the prediction parameter selection based on template matching quality. When template matching indicates good correlation, IntraTMP mode is used; when correlation is poor or noise is detected, the system switches to other prediction modes with different parameters, thus adapting prediction accuracy to the actual content characteristics
2Measurement precision
If multiple prediction modes are combined, then prediction accuracy is improved, but decoding complexity increases
Solution Approach 1:
The patent segments the prediction process into distinct stages: template matching stage, mode selection stage, and prediction execution stage. By dividing the overall prediction task into separable components with clear decision boundaries, the system can evaluate multiple modes without requiring full processing of all modes for every block
Solution Approach 2:
The patent performs partial prediction mode evaluation by using template matching as a preliminary filter. Only blocks that meet certain template matching criteria proceed to full multi-mode prediction evaluation, while others use the simpler IntraTMP approach, thus avoiding excessive computation for all blocks
Data Source
AI summary
Provided in the embodiments of the present application are a decoding method, a coding method, a decoder and a coder. The decoding method comprises: determining a residual block of the current block in the current sequence on the basis of a code stream; determining a first prediction block of the current block on the basis of an intra template matching prediction (IntraTMP) mode; determining a second prediction block of the current block on the basis of a first prediction mode, wherein the first prediction mode is different from the IntraTMP mode; determining a target prediction block of the current block on the basis of the first prediction block and the second prediction block; and obtaining a reconstructed block of the current block on the basis of the residual block of the current block and the target prediction block.