Intra Prediction Template Matching for Higher Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques face challenges in achieving high coding efficiency and image enhancement due to limitations in intra-prediction methods, particularly with restricted use of spatially adjacent pixels, necessitating improved intra-prediction approaches.
Innovation Solution
A video coding method and apparatus that utilizes template matching-based intra prediction by defining previously reconstructed pixels adjacent to the current block as a template and performing a search in the neighboring reconstructed region to select an optimal intra predictor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional intra-prediction methods using spatially adjacent pixels are used, then device complexity is reduced, but coding efficiency and prediction performance deteriorate
Solution Approach 1:
The patent performs preliminary template matching in the reconstructed region to identify similar patterns before generating the final prediction. By pre-processing the reconstructed pixels to create templates and performing similarity searches, the method prepares prediction candidates in advance, improving coding efficiency without significantly increasing overall system complexity
Solution Approach 2:
The patent creates template copies from reconstructed pixels and searches for matching templates in the reconstructed region. By copying and comparing template patterns rather than directly using complex mathematical transformations, the method achieves better prediction performance with manageable computational complexity
2Measurement precision
If multiple line buffers and advanced intra-prediction techniques (MRL, MIP) are used, then prediction performance is improved, but device complexity increases
Solution Approach 1:
The patent extracts template patterns from reconstructed pixels and performs matching based on these extracted features rather than using multiple line buffers. By taking out the essential pattern information and performing direct template matching, the method achieves comparable or better prediction accuracy without the complexity of multiple reference line buffers
Solution Approach 2:
The patent replaces the mechanical buffer system (multiple line buffers) with a template matching approach that operates directly on reconstructed pixel data. This substitution eliminates the need for complex buffer management while achieving improved prediction performance through pattern recognition
3Reliability
If template matching search in reconstructed region is performed, then prediction performance is enhanced, but computational complexity increases
Solution Approach 1:
The patent performs template matching locally within the reconstructed region adjacent to the current block, focusing computational effort on the most relevant areas. By limiting the search to locally reconstructed pixels rather than the entire picture, the method enhances prediction reliability while controlling computational complexity through spatial localization
Data Source
AI summary
A method and an apparatus for video coding using template matching-based intra prediction are disclosed. The video coding method and the video coding apparatus select an intra predictor by defining previously reconstructed pixels spatially adjacent to the current block as a template and performing a template matching-based search in the previously reconstructed region neighboring the current block.