Image Encoding Template Matching Merge List Reordering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image encoding/decoding technologies face challenges in improving the accuracy of motion information and motion prediction, particularly in constructing a merge list for a target block, which affects the efficiency of inter-prediction.
Innovation Solution
The method involves determining prediction information for a target block based on specific information and performing prediction using template matching to enhance the accuracy of motion vectors and reorder them efficiently for improved inter-prediction performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional inter-prediction technology is used for image encoding, then the encoding process can be performed with standard complexity, but the accuracy of motion information and motion prediction performance deteriorates
Solution Approach 1:
The patent applies preliminary action by constructing the merge list in advance using template matching and motion vector reordering before the actual prediction process. The encoder pre-processes and sorts motion vectors of candidate blocks according to their similarity with the target block's motion vector, so that when prediction is performed, the most accurate motion information is already positioned optimally in the merge list, eliminating the need for complex real-time adjustments during encoding.
Solution Approach 2:
The patent changes the parameter of motion vector ordering by reordering motion vectors based on template matching results. Instead of using conventional fixed ordering, the system dynamically adjusts the sequence of motion vectors in the merge list according to their correlation with the target block, using parameters such as template similarity metrics to determine the optimal arrangement, thereby improving motion prediction accuracy without proportionally increasing complexity.
2Measurement precision
If template matching is applied to improve motion vector accuracy, then the precision of motion prediction increases, but the computational complexity and processing time increases
Solution Approach 1:
The patent extracts and utilizes only the essential motion vector information and template matching results needed for effective prediction, rather than processing all possible candidate blocks exhaustively. By selecting and reordering only the most relevant motion vectors based on template similarity, the system achieves high accuracy while reducing the computational burden compared to evaluating all potential candidates in detail.
Solution Approach 2:
The patent applies partial action by performing template matching and motion vector reordering on a selective subset of candidate blocks rather than all possible candidates. The system processes enough candidates to achieve the desired prediction accuracy threshold, then stops, avoiding the excessive computation that would result from exhaustive evaluation of all potential merge candidates, thus balancing accuracy with processing time.
3Productivity
If motion vectors are reordered based on template matching, then the inter-prediction efficiency improves, but the complexity of the prediction process increases
Solution Approach 1:
The patent applies self-service by enabling the merge list construction process to automatically reorder motion vectors based on their inherent similarity to the target block's motion vector. The system uses the motion vectors themselves and template matching results to determine their own optimal arrangement without requiring external intervention or complex control logic, allowing the prediction process to self-optimize based on the data it processes.
Solution Approach 2:
The patent implements feedback by using template matching results to inform the reordering of motion vectors in the merge list. The similarity metrics obtained from template matching provide feedback about which motion vectors are most relevant, and this feedback is used to adjust the ordering of candidates, creating a closed-loop system where prediction accuracy continuously improves based on the performance information gathered during the process.
Data Source
AI summary
Disclosed therein are a method, an apparatus, and a storage medium for image encoding/decoding. Templating matching is used for prediction for a target block. Prediction information to be applied to encoding of the target block is determined, and prediction for the target block that uses the determined prediction information is performed. The prediction information includes a list and multiple candidates in the list. When the multiple candidates in the list are configured, template matching between blocks may be used. A candidate selected from among multiple candidates configured using template matching is used for prediction for the target block.