Motion Vector Predictor Selection for Video Coding Error Resistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional moving picture coding methods using inter prediction lack error resistance during coding and decoding processes.
Innovation Solution
A moving picture coding method that calculates and codes a motion vector predictor by selecting from a list of candidates, including predetermined fixed values, to improve error resistance by ensuring consistent predictor selection during both encoding and decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional inter prediction coding is used to reduce information amount, then coding efficiency is improved, but error resistance deteriorates
Solution Approach 1:
The patent applies preliminary action by pre-defining a fixed set of motion vector predictor candidates (including zero vectors and boundary-based predictors) before the actual prediction process. This predetermined candidate list ensures that both encoder and decoder have consistent fallback options, improving error resistance while maintaining coding efficiency through selective use of these pre-prepared predictors.
Solution Approach 2:
The patent changes the parameter of motion vector predictor selection by introducing a fixed candidate list with specific predefined values (such as zero vectors and boundary-derived vectors) rather than relying solely on dynamically calculated predictors. This parameter change ensures consistent behavior between encoder and decoder, enhancing error resistance without significantly impacting coding efficiency.
2Measurement precision
If motion vector predictor candidates are dynamically generated, then prediction accuracy is improved, but consistency between encoding and decoding deteriorates
Solution Approach 1:
The patent segments the motion vector predictor candidates into two distinct groups: dynamically generated candidates (for high accuracy) and fixed predefined candidates (for consistency). This segmentation allows the system to use dynamic predictors when available for improved accuracy, while falling back to fixed candidates to ensure consistency between encoding and decoding processes.
Solution Approach 2:
The patent provides beforehand cushioning by including fixed predefined candidates (such as zero vectors and boundary-based predictors) in the candidate list. These predetermined options act as a safety cushion that ensures both encoder and decoder have consistent fallback choices, preventing divergence even when dynamic prediction fails or produces inconsistent results.
Data Source
AI summary
A moving picture coding method for calculating a motion vector predictor to be used when coding a motion vector of a current block, and coding the current block includes: (S102) obtaining motion vector predictor candidates a fixed number of which is two or more; (S103) selecting the motion vector predictor to be used for coding the motion vector of the current block, from among the motion vector predictor candidates the fixed number of which is two or more; and (S104) coding an index for identifying the selected motion vector predictor according to the fixed number of two or more, wherein the motion vector predictor candidates the fixed number of which is two or more include one or more candidates each having a predetermined second fixed value.