Image Encoding Motion Vector Predictor Selection for Temporal Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion-vector-predictor specification modes do not effectively handle cases where a block is encoded with reference to a picture that temporally matches the block, leading to inefficient encoding efficiency.
Innovation Solution
An image encoding method that determines reference pictures for a current block, adds motion vectors of adjacent or spatially close blocks to a list, selects a motion vector predictor, and encodes the current block using the selected predictor, with the option to generate motion vectors for all blocks in a picture using reference pictures from a different view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional motion-vector-predictor specification mode is used, then motion vector prediction can be performed, but encoding efficiency deteriorates when block is encoded with reference to temporally matching picture
Solution Approach 1:
The patent segments the motion vector predictor candidate list into different groups based on the picture type and reference picture characteristics. For temporal matching pictures, it creates a separate handling mechanism that excludes certain candidates from the prediction list, thereby resolving the contradiction between maintaining prediction capability and ensuring encoding efficiency.
Solution Approach 2:
The patent applies different motion vector predictor selection rules to different local situations. When the current picture is a temporal matching picture, it uses one set of rules (excluding certain candidates), while for other picture types it uses different rules (allowing broader candidate selection), thus optimizing encoding efficiency for each specific case.
2Measurement precision
If motion vector predictor is selected from adjacent blocks, then prediction can be performed, but prediction accuracy deteriorates when reference picture is temporally matching
Solution Approach 1:
Instead of blindly including all adjacent blocks as predictors, the patent inverts the conventional approach by selectively excluding certain adjacent blocks when the reference picture is temporally matching. This inversion of the standard prediction procedure resolves the accuracy-efficiency contradiction by removing inappropriate candidates from the prediction list.
3Adaptability or versatility
If all blocks are encoded with reference to temporally matching picture, then encoding completeness is improved, but motion vector predictor generation becomes problematic
Solution Approach 1:
The patent introduces dynamic control in motion vector predictor generation based on picture type detection. When a temporal matching picture is detected, the system dynamically adjusts the predictor generation rules by excluding certain candidates, whereas for other picture types it follows standard generation procedures. This dynamic adaptation resolves the complexity issue while maintaining versatility.
Data Source
AI summary
The image encoding method includes: generating a reference picture list for determining reference pictures; determining a co-located picture; adding, to the list, a motion vector for each corresponding block that is included in a current picture to be encoded and is adjacent to a current block to be encoded or that is included in the co-located picture and close to the current block; selecting a motion vector predictor from the list; and encoding a difference between a motion vector of the current block encoded using the motion vector of the current block and the motion vector predictor selected. In the encoding, when the current picture is of a specific picture type, the motion vector is generated for each of all blocks included in the current picture with reference to the reference pictures belonging to a view different from a view to which the current picture belongs.