Disparity Vector Construction for 3D-HEVC
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multiview video coding techniques face challenges in determining disparity vectors efficiently, particularly in complex scenarios where depth maps are not readily available, leading to increased processing complexity and memory usage.
Innovation Solution
A method for determining disparity vectors based on motion information of spatial or temporal neighboring blocks, allowing for inter-view motion vector prediction and residual prediction without the need for estimated depth maps, using spatial disparity vectors, temporal disparity vectors, and implicit disparity vectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If depth maps are used for determining disparity vectors in multiview video coding, then the accuracy of inter-view prediction is improved, but the processing complexity and memory usage increase significantly
Solution Approach 1:
The patent extracts and utilizes motion information from neighboring blocks (both spatial and temporal) to determine disparity vectors, eliminating the need for complex depth map generation and processing. By taking out the essential motion cues from available neighboring block data, the method achieves accurate disparity vector determination without the computational burden of depth maps
Solution Approach 2:
The patent copies motion vector information from spatial and temporal neighboring blocks to construct disparity vectors. Instead of generating new depth information, it replicates and adapts existing motion information from adjacent blocks, significantly reducing processing complexity while maintaining prediction accuracy
2Measurement precision
If depth maps are generated and processed for disparity vector determination, then inter-view prediction accuracy is improved, but memory usage increases
Solution Approach 1:
The method extracts disparity vector information directly from motion information of neighboring blocks that are already present in the decoded video stream, eliminating the need to store and process separate depth map data structures. This extraction approach achieves the same predictive accuracy without the additional memory burden of depth maps
3Quantity of substance
If complex algorithms are used to determine disparity vectors without depth maps, then memory usage is reduced, but processing complexity increases
Solution Approach 1:
The patent employs a copying strategy where motion vector information from spatial and temporal neighboring blocks is directly reused and adapted to form disparity vectors. This simple copying and adaptation mechanism avoids complex algorithms while efficiently determining accurate disparity vectors with minimal memory requirements
Data Source
AI summary
When coding multiview video data, a video coder can code one or more pictures in one or more reference views, including a first reference view and determine a disparity vector for a current block based on motion information of one or more neighboring blocks of the current block, wherein the current block is in a second view, wherein the disparity vector points from the current block to a corresponding block in a picture of the same time instance in one of the one or more reference views.