IMPROVED MULTISTAGE LEARNING FOR MULTIBODY MOTION SEGMENTATION

Yasuyuki Sugaya, Kenichi Kanatani

2010

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.

Paper Citation

in Harvard Style

Sugaya Y. and Kanatani K. (2010). IMPROVED MULTISTAGE LEARNING FOR MULTIBODY MOTION SEGMENTATION . In Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2010) ISBN 978-989-674-028-3, pages 199-206. DOI: 10.5220/0002814601990206

in Bibtex Style

@conference{visapp10,
author={Yasuyuki Sugaya and Kenichi Kanatani},
title={IMPROVED MULTISTAGE LEARNING FOR MULTIBODY MOTION SEGMENTATION},
booktitle={Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2010)},
year={2010},
pages={199-206},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002814601990206},
isbn={978-989-674-028-3},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2010)
TI - IMPROVED MULTISTAGE LEARNING FOR MULTIBODY MOTION SEGMENTATION
SN - 978-989-674-028-3
AU - Sugaya Y.
AU - Kanatani K.
PY - 2010
SP - 199
EP - 206
DO - 10.5220/0002814601990206