Inter-Prediction MVP Candidate Reordering for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-resolution and high-quality video data requires efficient compression techniques to reduce transmission and storage costs, as existing methods struggle to effectively manage the increased information volume.
Innovation Solution
A method and apparatus for video coding that derive and reorder motion vector predictor (MVP) candidate lists based on costs, considering both luma and chroma components, to enhance prediction accuracy and efficiency, and include a refine MVP candidate for improved coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional inter prediction methods are used for high-resolution video, then transmission and storage costs increase due to increased bit amounts, but simplifying the prediction method reduces prediction accuracy
Solution Approach 1:
The patent changes the parameter of MVP candidate selection by deriving multiple MVP candidates from different neighboring blocks (A0, A1, B0, B1, B2) and ordering them based on cost calculations. This allows the system to select the most appropriate candidate for each current block, improving prediction accuracy while maintaining efficient bit usage through targeted refinement only when necessary.
Solution Approach 2:
The patent segments the prediction process into multiple stages: initial MVP candidate derivation from various neighboring blocks, cost-based ordering of candidates, selection of best candidate, and optional refinement. This segmentation allows the system to allocate computational resources efficiently, performing detailed refinement only for blocks where it provides the most benefit, thereby reducing overall bit amounts while maintaining prediction accuracy.
2Measurement precision
If multiple MVP candidates are derived and reordered based on cost, then prediction accuracy improves, but computational complexity increases
Solution Approach 1:
The patent applies partial action by performing refinement processing selectively rather than for all current blocks. The refinement is applied only when it provides significant improvement, determined through cost calculations. This approach achieves high prediction accuracy for blocks where it matters most while avoiding unnecessary computational complexity for blocks where simple MVP selection suffices.
Solution Approach 2:
The patent changes the parameter of candidate evaluation by introducing cost calculations that consider both prediction accuracy and bit amount. This allows the system to order MVP candidates based on a composite metric that balances accuracy improvement against computational cost, enabling efficient selection of the most beneficial candidates without exhaustive processing of all possibilities.
3Productivity
If refinement processing is applied to MVP candidates, then coding efficiency improves, but processing time increases
Solution Approach 1:
The patent applies refinement processing partially, only to MVP candidates that show potential for improvement based on cost calculations. By evaluating candidates in ordered sequence and applying refinement only when beneficial, the system achieves coding efficiency improvements for the most critical blocks while limiting processing time expenditure to only where it provides measurable benefit.
Solution Approach 2:
The patent performs preliminary cost calculations and candidate ordering before applying refinement processing. This preliminary action identifies which candidates are worth refining, allowing the system to avoid wasting processing time on candidates that would not benefit from refinement. The preliminary ordering based on cost metrics guides the subsequent refinement process to focus only on promising candidates.
Data Source
AI summary
A video decoding method performed by a decoding apparatus includes obtaining information regarding inter-prediction of a current block via a bitstream; forming an MVP candidate list based on neighbouring blocks of the current block; deriving costs with respect to MVP candidates that are included in the MVP candidate list; deriving a revised MVP candidate list based on the costs with respect to the MVP candidates; deriving MVP of the current block based on the revised MVP candidate list; deriving a motion vector of the current block based on the MVP and MVD of the current block; and performing prediction on the current block based on a reference picture index and the motion vector of the current block. The information regarding inter-prediction includes the reference picture index and the MVD.