Video Decoding Motion Refinement Using Reference Picture Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion information refinement methods in video decoding have poor refinement effects, leading to inaccurate prediction and affecting decoding performance.
Innovation Solution
A video decoding method that refines motion information based on reference pictures to improve decoding accuracy, involving determining first motion information, refining it to obtain second motion information, and using this to calculate a prediction value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current motion information refinement methods are used, then decoding process is simple, but refinement effect is poor and prediction accuracy is insufficient
Solution Approach 1:
The patent segments the current block into multiple sub-blocks and performs motion information refinement separately for each sub-block. This segmentation allows the decoder to achieve better refinement accuracy by processing smaller regions independently, while the modular approach keeps the implementation complexity manageable through systematic organization of the refinement process.
Solution Approach 2:
The patent applies different motion information refinement strategies to different sub-blocks based on their local characteristics. By determining motion information for each sub-block independently and using local reference picture information, the system achieves locally optimized refinement quality without requiring complex global processing, thus balancing accuracy and complexity.
2Reliability
If motion information is not refined, then decoding complexity is low, but prediction accuracy is insufficient and decoding performance is affected
Solution Approach 1:
The patent performs motion information refinement as a preliminary step before generating the final prediction for the current block. By refining motion information in advance using reference picture data and sub-block segmentation, the system ensures high prediction accuracy is achieved before the main decoding process, improving reliability while keeping the refinement complexity isolated and manageable.
Solution Approach 2:
The decoder performs motion information refinement using its own internal resources and already-decoded reference pictures, without requiring additional external information or complex external processing. This self-service approach improves decoding performance through enhanced accuracy while avoiding the complexity of external systems or additional hardware components.
Data Source
AI summary
A video decoding method includes: determining first motion information of a current block; refining the first motion information based on motion information of a reference picture of the current block to obtain second motion information of the current block; and determining a prediction value of the current block based on the second motion information.