Motion Candidate Derivation via Template Matching in Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently deriving motion information for video blocks, which affects compression efficiency and bitstream size.
Innovation Solution
The method involves determining a set of motion candidates for a current video block, refining these candidates through local searches based on a template matching cost rule, and selecting a target motion candidate for the conversion process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion information is signaled in the bitstream, then decoding accuracy is improved, but bitstream size increases
Solution Approach 1:
The patent uses template matching to create a reference template from the current video block, then copies and compares this template with corresponding blocks in reference frames to derive motion information. This copying approach enables accurate motion estimation without requiring extensive motion information to be signaled in the bitstream.
Solution Approach 2:
The system performs self-service by deriving its own motion information through template matching operations. The video decoder independently calculates motion vectors by comparing the template with reference frame blocks, eliminating the need to receive pre-encoded motion information from the bitstream.
2Measurement precision
If template matching is performed with large search range, then motion candidate accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the search process by first performing template matching to identify potential motion candidates, then performing refinement searches only around these candidates rather than across the entire search range. This segmentation reduces computational complexity while maintaining accuracy.
Solution Approach 2:
The patent performs preliminary template matching to identify promising motion candidates before conducting refinement searches. This preliminary action narrows down the search space, allowing the system to achieve high accuracy without exhaustively searching the entire range.
3Quantity of substance
If motion information is derived at the decoder, then bitstream compression is improved, but decoding complexity increases
Solution Approach 1:
The decoder derives motion information by copying and comparing templates from the current block with corresponding regions in reference frames. This copying-based approach enables compression without motion information signaling while keeping the derivation process relatively simple and efficient.
Solution Approach 2:
The patent changes the parameter of motion information representation by deriving motion vectors through template matching cost calculations rather than receiving them directly. This parameter change enables compression while the systematic derivation process manages decoding complexity.
Data Source
AI summary
Embodiments of the present disclosure provide methods, apparatuses and computer storage media for video processing. One example method comprises determining, during a conversion between a current video block of a video and a bitstream of the video, a motion candidate for the current video block; refining the motion candidate by performing a local search around the motion candidate based on a template matching cost rule; and performing the conversion based on the refined motion candidate, wherein a search order of the local search is related to a coding mode of the current video block.