Epipolar Geometry Motion Estimation for Multi-View Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-view video compression schemes face high computational costs due to inter-viewpoint prediction, particularly in convergent multi-view camera setups, where traditional motion estimation algorithms are inefficient in tracking large and irregular motion, leading to reduced coding efficiency.
Innovation Solution
The use of epipolar geometry-based fast motion estimation, which computes an epipolar line and determines a search starting point for motion estimation, allowing for a reduced search space and efficient motion vector calculation, incorporating local diamond and unsymmetrical rood-pattern searches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional full-search motion estimation is used to track large and irregular motion in convergent multi-view camera setups, then motion prediction performance is maintained, but computational cost increases significantly
Solution Approach 1:
The patent changes the search strategy parameters by using epipolar geometry to define constrained search paths and adaptive search patterns. Instead of exhaustive full-search, the method adjusts search parameters (search range, search pattern, starting points) based on epipolar constraints and motion characteristics, achieving reliable motion tracking with reduced computational complexity
Solution Approach 2:
The motion estimation process is segmented into multiple stages: coarse search using epipolar constraints to locate approximate motion vectors, followed by refined search using adaptive patterns. This segmentation allows the system to achieve accurate motion prediction without the computational burden of uniform full-search across all blocks
2Measurement precision
If traditional full-search motion estimation is used, then motion tracking accuracy is maintained, but processing time increases
Solution Approach 1:
The patent performs preliminary action by using epipolar geometry to pre-determine constrained search paths and identify likely motion vector ranges before conducting the actual motion estimation. This preliminary constraint setup enables accurate motion tracking while significantly reducing the search space and processing time required
Solution Approach 2:
The search strategy is made dynamic by adapting search patterns and ranges based on local motion characteristics and epipolar constraints. The system dynamically adjusts search intensity and methodology for different macroblocks, maintaining high tracking accuracy while optimizing processing time through selective detailed search only where necessary
3Productivity
If fast motion estimation algorithms are used to reduce computational load, then processing speed increases, but they are inefficient for inter-viewpoint prediction in multi-view video
Solution Approach 1:
The patent applies local quality by using different search strategies for different regions and motion types within the video sequence. Fast search methods are applied where appropriate, while more thorough search is performed where motion complexity requires it, all within the epipolar constraint framework. This localized adaptation maintains prediction performance while achieving processing speed improvements
Data Source
AI summary
Described is fast motion estimation based upon epipolar geometry, which can be used in compressing multi-view video. An epipolar line is computed based on a point (e.g., a centroid point) in a macroblock to be predicted, and a temporary starting point in an image is determined, such as a median predicted search center. A search starting point is further determined based on the temporary starting point and the epipolar line, e.g., a point on the epipolar line corresponding to an intersecting line that is projected orthogonally from the temporary point to the epipolar line. A motion estimation mechanism searches the search space to produce a motion vector. The search may be conducted starting at the search starting point in a reduced search area located around the epipolar line, e.g., a local diamond search and/or rotated unsymmetrical rood-pattern search.