Region-Based Template Matching for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face increased decoder run-time due to the computational complexity of template matching methods used for decoder-side motion vector derivation in video compression.
Innovation Solution
The implementation of region-based template matching (RTM) for inter frame coding, which partitions the search window into multiple regions, allowing for independent prediction signals and reducing the search area, thereby decreasing computational complexity and improving coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If template matching methods are used for decoder-side motion vector derivation, then coding efficiency is improved, but decoder run-time increases
Solution Approach 1:
The patent divides the search window into multiple regions (e.g., first region, second region, third region) and performs template matching independently in each region to find multiple candidate predictor blocks. This segmentation reduces the computational complexity of searching the entire window while maintaining coding efficiency through multiple candidates.
Solution Approach 2:
Instead of performing exhaustive template matching across the entire search window, the patent performs partial matching by limiting the search to specific regions within the window. This partial action approach reduces decoder run-time while still providing sufficient candidate blocks for accurate motion compensation.
2Measurement precision
If the search window is large, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The search window is segmented into multiple regions, allowing the patent to maintain a large overall search window for accurate prediction while reducing computational complexity by processing smaller regions independently. Each region contributes candidate blocks that collectively provide accurate prediction without requiring exhaustive search of the entire window.
Solution Approach 2:
Different regions within the search window can be processed with different levels of detail or search strategies. The patent applies template matching locally in each region to find candidate blocks, allowing local optimization that balances prediction accuracy with computational efficiency for each specific area.
Data Source
AI summary
Video decoder and/or video encoder, configured to determine a set of search area location candidates in a reference picture of a video; match the set of search area location candidates with a current template area adjacent to a current block of a current picture to obtain a best matching search area location candidate; select, out of a search area positioned in the reference picture at the best matching search area location candidate, a set of one or more predictor blocks by matching the current template area against the search area; and predictively decode/encode the current block from/into a data stream based on the set of one or more predictor blocks.