View Synthesis Prediction Signaling for 3D Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 3D video coding techniques face inefficiencies in view synthesis prediction due to legacy hardware limitations and computational overhead, particularly when generating view synthesis pictures on the fly, which impacts coding and computational efficiency.
Innovation Solution
The proposed solution involves signaling a reference index flag at the macroblock or MB partition level to indicate whether view synthesis prediction is applied, allowing for flexible reference index usage and avoiding the need for pre-generated view synthesis pictures, thus optimizing computational and memory efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If view synthesis prediction is applied using conventional techniques, then prediction accuracy is improved, but computational overhead and processing time increase significantly
Solution Approach 1:
The patent pre-generates view synthesis pictures and stores them in reference picture lists before they are needed for prediction. This preliminary action eliminates the need for real-time synthesis during decoding, significantly reducing processing time while maintaining prediction accuracy. The reference pictures are prepared in advance and can be directly used for inter-view prediction.
Solution Approach 2:
The patent extracts the computationally intensive view synthesis operation from the real-time prediction process and separates it into a pre-processing stage. By taking out the synthesis operation and performing it beforehand, the system avoids the computational overhead during actual video decoding while preserving the accuracy benefits of view synthesis prediction.
2Productivity
If view synthesis pictures are pre-generated and stored, then coding efficiency is improved, but memory requirements and device complexity increase
Solution Approach 1:
The patent makes the reference picture list structure universal by allowing it to serve multiple functions: storing both conventional temporal reference pictures and pre-generated view synthesis pictures. This multi-functionality enables the system to improve coding efficiency through view synthesis while reusing existing hardware structures without significant complexity increases.
Solution Approach 2:
The patent creates simplified copies of view synthesis pictures and stores them in the reference picture list. These copies are sufficient for prediction purposes and can be generated using optimized algorithms that reduce the computational and memory burden compared to storing full-resolution original views, thereby improving coding efficiency while limiting complexity growth.
3Adaptability or versatility
If reference picture lists are modified to include view synthesis pictures, then prediction flexibility is improved, but compatibility with legacy hardware decreases
Solution Approach 1:
The patent implements a dynamic reference picture list management system where the inclusion of view synthesis pictures is controlled by a flag or mode indicator. This allows the system to adaptively switch between conventional and enhanced prediction modes, providing prediction flexibility when needed while maintaining compatibility with legacy hardware that cannot process view synthesis pictures by simply not including them in the reference list.
Data Source
AI summary
In an example, a method of decoding video data includes determining whether a reference index for a current block corresponds to an inter-view reference picture, and when the reference index for the current block corresponds to the inter-view reference picture, obtaining, from an encoded bitstream, data indicating a view synthesis prediction (VSP) mode of the current block, where the VSP mode for the reference index indicates whether the current block is predicted with view synthesis prediction from the inter-view reference picture.