Authors:
Yasuyuki Sugaya
1
and
Kenichi Kanatani
2
Affiliations:
1
Toyohashi University of Technology, Japan
;
2
Okayama University, Japan
Keyword(s):
Multibody motion segmentation, GCAP, Taubin method, EM algorithm, Multistage optimization.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Computer Vision, Visualization and Computer Graphics
;
Human-Computer Interaction
;
Image and Video Analysis
;
Methodologies and Methods
;
Motion and Tracking
;
Motion, Tracking and Stereo Vision
;
Pattern Recognition
;
Physiological Computing Systems
;
Segment Cluster Tracking
;
Software Engineering
;
Video Analysis
Abstract:
We present an improved version of the MSL method of Sugaya and Kanatani for multibody motion segmentation. We replace their initial segmentation based on heuristic clustering by an analytical computation based on GPCA, fitting two mbox 2-D affine spaces in mbox 3-D by the Taubin method. This initial segmentation alone can segment most of the motions in natural scenes fairly correctly, and the result is successively optimized by the EM algorithm in mbox 3-D , mbox 5-D , and mbox 7-D . Using simulated and real videos, we demonstrate that our method outperforms the previous MSL and other existing methods. We also illustrate its mechanism by our visualization technique.