Template-Based Video Block Prediction With Filtered Reference Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently utilizing spatial and temporal redundancy for improved compression, particularly in intra and inter prediction methods, leading to suboptimal video encoding and decoding performance.
Innovation Solution
The implementation of template-based prediction methods, including intra and inter prediction modes, where filtering is applied to current and reference templates to determine a linear model for reconstructing blocks, using different filters for each template and signaling filtering flags at block or higher levels, enhancing the prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filtering is applied to templates to improve prediction accuracy, then prediction quality improves, but computational complexity increases
Solution Approach 1:
The patent applies filtering operations to modify template samples before linear model determination, changing the parameters of the prediction process. By filtering samples in the current and reference templates, the patent improves prediction accuracy while managing computational complexity through selective application of filtering operations.
2Measurement precision
If different filters are applied to current and reference templates to improve prediction precision, then prediction quality improves, but processing complexity increases
Solution Approach 1:
The patent applies different filtering operations to different parts of the prediction process - specifically, different filters are applied to current template samples versus reference template samples. This local differentiation allows optimized prediction precision for each template type while managing overall processing complexity through targeted filtering rather than uniform processing.
3Productivity
If template matching prediction is used to compress video data, then compression efficiency improves, but reconstruction quality may degrade
Solution Approach 1:
The patent uses a linear model determined from filtered template samples to predict current block samples based on reference block samples. This feedback mechanism allows the system to maintain reconstruction quality by continuously adapting predictions based on the relationship between current and reference templates, while preserving compression efficiency through predictive coding.
Data Source
AI summary
Aspects of the disclosure includes methods and apparatuses for video decoding and video encoding and a method of processing visual media data. Coded information in a bitstream is received. The coded information indicates whether filtering is to be applied to at least one of a current template of a current block in a current picture and a reference template of a reference block. The current block is predicted based on the reference block. When the coded information indicates that the filtering is to be applied, a plurality of samples in the at least one of the current template and the reference template is filtered. A linear model between the current template and the reference template is determined based on the filtered plurality of samples. The current block is reconstructed based on the linear model and the reference block.