Stereo matching using M-estimators

  • Christian Menard
  • Aleš Leonardis
Stereo and Correspondence
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1296)


Stereo computation is just one of the vision problems where the presence of outliers cannot be neglected. Most standard algorithms make unrealistic assumptions about noise distributions, which leads to erroneous results that cannot be corrected in subsequent processing stages. In this work the standard area-based correlation approach is modified so that it can tolerate a significant number of outliers. The approach exhibits a robust behaviour not only in the presence of mismatches but also in the case of depth discontinuities. Experimental results are given on synthetic and real images.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    P. J. Besl, J. B. Birch, and L. T. Watson. Robust window operators. In Proceedings of the 2nd ICCV, pages 591–600. IEEE, Dec 1988.Google Scholar
  2. 2.
    D.N. Bhat and S.K. Nayar. Ordinal measures for visual correspondence. Technical Report CUCS-009-96, Columbia University,Dept. of Computer Science, February 1996.Google Scholar
  3. 3.
    H. Helmke, R. Janssen, and G. Saur. Automatische Erzeugung dreidimensionaler Kantenmodelle aus mehreren zweidimensionalen Objektansichten. In Grabkopf R.E., editor, Mustererkennung 1990, 12. DAGM Symposium, pages 617–624, 1990.Google Scholar
  4. 4.
    P. J. Huber. Robust Statistics. Wiley, New York, 1981.Google Scholar
  5. 5.
    M.R.M. Jenkin, A.D. Jepson, and J.K. Tsotsos. Techniques for disparity measurements. CVGIP, 53:14–30, 1991.CrossRefGoogle Scholar
  6. 6.
    D. Y. Kim, J. J. Kim, P. Meer, D. Mintz, and A. Rosenfeld. Robust computer vision: A least-median of squares based approach. In Proceedings of the Image Understanding Workshop, pages 1117–1134, Palo Alto, CA, May 1989. DARPA.Google Scholar
  7. 7.
    W. Luo and H. Maitre. Using surface model to correct and fit disparity data in stereo vision. In Tenth International Conference on Pattern Recognition, pages 60–64, Atlantic City, NJ, June 16–21 1990.Google Scholar
  8. 8.
    D. Marr and T. Poggio. A computational theory of human stereo vision. Proc. R. Soc. Lond. B., 204:301–328, 1979.PubMedGoogle Scholar
  9. 9.
    P. Meer, D. Mintz, A. Rosenfeld, and D. Y. Kim. Robust regression methods for computer vision: A review. International Journal of Computer Vision, 6(1):59–70, 1991.CrossRefGoogle Scholar
  10. 10.
    C. Menard. Robust Stereo and Adaptive Matching in Correlation Scale-Space. PhD thesis, TU Wien, Institut für Automation, PRIP, Wien, 1996.Google Scholar
  11. 11.
    Azriel Rosenfeld and Avinash C. Kak. Digital Picture Processing Volume 2. Academic Press, Inc., 1982.Google Scholar
  12. 12.
    P. J. Rousseuw and A. M. Leroy. Robust Regression and Outlier Detection. Wiley, New York, 1987.Google Scholar
  13. 13.
    B. G. Schunck. Robust computational vision. In Proc. Of the IWRCV, Seattle, WA, Oct 1990.Google Scholar
  14. 14.
    S. S. Sinha and B. G. Schuck. A two-stage algorithm for discontinuity-preserving surface reconstruction. IEEE Trans. on PAMI, 14(1):36–55, Jan 1992.Google Scholar
  15. 15.
    J. Subrahmonia, J. Hung, and D.B. Cooper. Model-based segmentation and estimation of 3d surfaces from two or more intensity images using markov random fields. IEEE Conf. on Pattern Recognition, 1:390–397, 1990.CrossRefGoogle Scholar
  16. 16.
    D. Terzopoulos, A. Witkin, and M. Kass. Stereo matching as constrained optimization using scale continuation methods. In Opt. Dig. PR/ SPIE 754, pages 92–99, 1987.Google Scholar
  17. 17.
    J.W. Tukey. Some advanced thoughts on the data analysis involved in configural polysampling directed toward high performance estimates. Technical Report 189, Series 2, Department of Statistics, Princeton University, Princeton,N.J., 1981.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Christian Menard
    • 1
  • Aleš Leonardis
    • 1
    • 2
  1. 1.Pattern Recognition and Image Processing GroupVienna University of TechnologyViennaAustria
  2. 2.Faculty of Computer and Information ScienceUniversity of LjubljanaLjubljana

Personalised recommendations