Motion segmentation by subspace separation and model selection
- 1 July 2001
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 586-591
- https://doi.org/10.1109/iccv.2001.937679
Abstract
Reformulating the Costeira-Kanade algorithm as a pure mathematical theorem independent of the Tomasi-Kanade factorization, we present a robust segmentation algorithm by incorporating such techniques as dimension correction, model selection using the geometric AIC, and least-median fitting. Doing numerical simulations, we demonstrate that oar algorithm dramatically outperforms existing methods. It does not involve any parameters which need to be adjusted empirically.Keywords
This publication has 11 references indexed in Scilit:
- Geotensity Constraint for 3D Surface Reconstruction under Multiple Light SourcesPublished by Springer Nature ,2000
- Motion segmentation based on factorization method and discriminant criterionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1999
- Geometric motion segmentation and model selectionPhilosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1998
- Geometric Information Criterion for Model SelectionInternational Journal of Computer Vision, 1998
- Multibody Grouping from Motion ImagesInternational Journal of Computer Vision, 1998
- A paraperspective factorization method for shape and motion recoveryPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1997
- Shape and motion from image streams under orthography: a factorization methodInternational Journal of Computer Vision, 1992
- Robust regression methods for computer vision: A reviewInternational Journal of Computer Vision, 1991
- Robust Regression and Outlier DetectionPublished by Wiley ,1987
- A Threshold Selection Method from Gray-Level HistogramsIEEE Transactions on Systems, Man, and Cybernetics, 1979