IMPROVED MULTISTAGE LEARNING FOR MULTIBODY MOTION SEGMENTATION

Yasuyuki Sugaya, Kenichi Kanatani

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.

References

  1. Costeira, J. P. and Kanade, T. (1998). A multibody factorization method for independently moving objects, Int. J. Computer Vision, 29, 159-179.
  2. Fan, Z., Zhou, J. and Wu, Y. (2006). Multibody grouping by inference of multiple subspace from highdimensional data using oriented-frames, IEEE Trans Patt. Anal. Mach. Intell., 28, 91-105.
  3. Gear, C. W. (1998). Multibody grouping from motion images, Int. J. Comput. Vision, 29, 133-150.
  4. Ichimura, N. (1999). Motion segmentation based on factorization method and discriminant criterion, Proc. 7th Int. Conf. Comput. Vis., Vol. 1, Kerkyra, Greece, 600- 605.
  5. Ichimura, N. (2000). Motion segmentation using feature selection and subspace method based on shape space, Proc. 15th Int. Conf. Pattern Recog., Vol. 3, Barcelona, 858-864.
  6. Inoue, K. and Urahama, K. (2001). Separation of multiple objects in motion images by clustering, Proc. 8th Int. Conf. Comput. Vis., Vol. 1, Vancouver, 219-224.
  7. Kanatani, K. (2002a). Evaluation and selection of models for motion segmentation, Proc. 7th Euro. Conf. Comput. Vis., Vol. 3, Copenhagen, 335-349.
  8. Kanatani, K. (2001). Motion segmentation by subspace separation and model selection, Proc. 8th Int. Conf. Comput. Vis., Vol. 2, Vancouver, 301-306.
  9. Kanatani, K. (2002b). Motion segmentation by subspace separation: Model selection and reliability evaluation, Int. J. Image Graphics, 2, 179-197.
  10. Kanatani, K. (1996). Statistical Optimization for Geometric Computation: Theory and Practice, Amsterdam: Elsevier. Reprinted (2005) New York: Dover.
  11. Kanatani, K. (2008). Statistical optimization for geometric fitting: Theoretical accuracy analysis and high order error analysis, Int. J. Comput. Vision, 80, 167-188.
  12. Kanatani, K. and Sugaya, Y. (2007). Performance evaluation of iterative geometric fitting algorithms, Comp. Stat. Data Anal., 52, 1208-1222.
  13. Poelman, C. J. and Kanade, T. (1997). A paraperspective factorization method for shape and motion recovery, IEEE Trans. Pattern Anal. Mach. Intell., 19, 206-218.
  14. Rao, S. R., Tron, R., Vidal R. and Ma, Y. (2008). Motion segmentation via robust subspace separation in the presence of outlying, incomplete, or corrupted trajectories, Proc. IEEE Conf. Comput. Vision Patt. Recog., Anchorage, AK.
  15. Schindler, K., Suter, D. and Wang, H. A model-selection framework for multibody structure-and-motion of image sequences, Int. J. Comput. Vision, 79, 159-177.
  16. Sugaya, Y. and Kanatani, Y. (2004). Multi-stage optimization for multi-body motion segmentation. IEICE Trans. Inf. & Syst., E87-D, 1935-1942.
  17. Taubin, G. (1991). Estimation of planer curves, surfaces, and non-planar space curves defined by implicit equations with applications to edge and range image segmentation, IEEE Trans. Patt. Anal. Mach. Intell., 13, 1115-1138.
  18. Tomasi, C. and Kanade, T. (1992). Shape and motion from image streams under orthography-A factorization method, Int. J. Comput. Vision, 9, 137-154.
  19. Tron, R. and Vidal, R. (2007). A benchmark for the comparison of 3-D motion segmentation algorithms, Proc. IEEE Conf. Comput. Vision Patt. Recog., Minneapolis, MN.
  20. Vidal, R., Ma, Y. and Sastry, S. (2005). Generalized principal component analysis (GPCA), IEEE Trans. Patt. Anal. Mach. Intell., 27, 1945-1959.
  21. Vidal, R. Tron, R. and Hartley, R. (2008). Multiframe motion segmentation with missing data using PowerFactorization and GPCA, Int. J. Comput. Vision, 79, 85- 105.
  22. Wu, Y. Zhang, Z., Huang, T. S. and Lin, J. Y. (2001). Multibody grouping via orthogonal subspace decomposition, sequences under affine projection, Proc. IEEE Conf. Computer Vision Pattern Recog., Vol. 2, Kauai, HI, 695-701.
  23. Yan, J. and Pollefeys, M. (2006). A general framework for motion segmentation: Independent, articulate, rigid, non-rigid, degenerate and nondegenerate, Proc. 9th Euro. Conf. Comput. Vision., Vol. 4, Graz, Austria, 94-104.
Download


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