Hardware-Friendly Template Matching Motion Vector Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The conventional template matching (TM) motion vector refinement process in video coding systems suffers from latency issues due to the need to wait for the refinement of previous motion vectors, leading to slow coding performance and increased latency in accessing reference data.
Innovation Solution
The proposed solution involves using a non-refined motion vector corresponding to a neighbouring CU to reconstruct the motion vector candidate list, allowing for faster MV derivation without waiting for the refinement process to complete, and performing motion compensation and TM refinement in parallel, with adjustments based on gradient values and bounding box restrictions to reduce latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the conventional template matching motion vector refinement process waits for the refinement of previous motion vectors, then the accuracy of motion vector refinement is improved, but the coding latency increases and processing speed decreases
Solution Approach 1:
The patent performs motion compensation using the initial motion vector before the template matching refinement is complete. By executing the motion compensation operation in advance with the available initial MV, the system avoids waiting for the refinement process, thereby reducing coding latency while maintaining the option to update the MV later with refined values
2Device complexity
If the system performs sequential processing of motion compensation and template matching refinement, then the implementation complexity is reduced, but the processing speed and throughput decrease
Solution Approach 1:
The patent initiates motion compensation operations before the template matching refinement completes by initiating the refinement process, using the initial motion vector for the actual compensation. This preliminary action allows overlapping of operations in time, improving throughput without significantly increasing hardware complexity
Solution Approach 2:
The patent introduces an intermediary initial motion vector that serves as a temporary substitute for the refined motion vector during the refinement process. This intermediary allows the motion compensation to proceed independently of the refinement timeline, enabling parallel or overlapped execution of operations
Data Source
AI summary
A method and apparatus for video coding system that utilizes low-latency template-matching motion-vector refinement are disclosed. According to this method, input data associated with a current block of a video unit in a current picture are received. Motion compensation is then applied to the current block according to an initial motion vector (MV) to obtain initial motion-compensated predictors of the current. After applying the motion compensation to the current block, template-matching MV refinement is applied to the current block to obtain a refined MV for the current block. The current block is then encoded or decoded using information including the refined MV. The method may further comprise determining gradient values of the initial motion-compensated predictors. The initial motion-compensated predictors can be adjusted by taking into consideration of the gradient values and/or MV difference between the refined and initial MVs.