Precise matching by Robust estimation of deformation and local coherence

  • Zhong-Dan Lan
  • Roger Mohr
  • Long Quan
Poster Session I
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1351)


In this paper, we present a linear method that incorporates information from neighboring pixels for sub-pixel matching. Two algorithms are presented. Both rely on a rough initial estimate of the disparity. The first one is optimized for pairs of images requiring negligible window deformation. The second method is slower but more general and more precise. It is applicable for large window deformation and eliminates false initial matches using robust estimation of the local affine window transformation. The first algorithm attains a precision of 0.05 pixels for interest points and 0.06 for random points in the translational case. For general case, if the deformation is small, the second method gives an accuracy of 0.05 pixels; while for large deformation, it gives an accuracy of about 0.06 pixels for points of interest and 0.10 pixels for random points.


Correlation Precision Robustness Convolution Matching 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    F. Ackermann. Digital image correlation: performance and potential application in photogrammetry. Photogrammetric Record, 64(11):429–439, October 1984.Google Scholar
  2. 2.
    P. Aschwanden and W. Guggenbiihl. Experimental results from a comparative study on correlation-type registration algorithms. In Forstner and Ruwiedel, editors, Robust Computer Vision, pages 268–282. Wichmann, 1992.Google Scholar
  3. 3.
    D. N. Bhat and S. K. Nayar. Ordinal measure for visual correspondence. In Proceedings of the Conference on Computer Vision and Pattern Recognition, San Francisco, California, USA, pages 351–357, San Francisco, California, June 1996.Google Scholar
  4. 4.
    W. Forstner. Mathematical models, accuracy aspects and quality control. International Archives of Photogrammetry, 24(3):176–189, June 1982.Google Scholar
  5. 5.
    D. Geiger, B. Ladendorf, and A. Yuille. Occlusions and binocular stereo. In G. Sandini, editor, Proceedings of the 2nd European Conference on Computer Vision, Santa Margherita Ligure, Italy, pages 425–433. Springer Verlag, 1992.Google Scholar
  6. 6.
    A.W. Gruen. Adaptative least squares correlation: a powerful image matching technique. S. Afr. Journal of Photogrammetry, Remote Sensing and Cartography, 14(3):175–187, 1985.Google Scholar
  7. 7.
    S.S. Intille and A.F. Bobick. Disparity-space images and large occlusion stereo. In Proceedings of the 3rd European Conference on Computer Vision, Stockholm, Sweden, pages 179–186. Springer-Verlag, 1994.Google Scholar
  8. 8.
    Z. D. Lan, R. Mohr, and P. Remagnino. Robust matching by partial correlation. In Proceedings of the sixth British Machine Vision Conference, Birmingham, England, pages 651–660, September 1995.Google Scholar
  9. 9.
    Z.D. Lan. Méthodes robustes en vision: application aux appariements visuels. Ph.d thesis, Institut National Polytechnique de Grenoble, 1997.Google Scholar
  10. 10.
    Z.D. Lan and R. Mohr. Robust location based partial correlation. Inria Rhone Alpes, January 1997. accepted by CAIP97.Google Scholar
  11. 11.
    J.L. Lotti and G. Giraudon. Adaptive window algorithm for aerial image stereo. In Proceedings of the 12th International Conference on Pattern Recognition, Jerusalem, Israel, pages 701–703, October 1994.Google Scholar
  12. 12.
    D. Marr. Vision. W.H. Freeman and Company, San Francisco, California, USA, 1982.Google Scholar
  13. 13.
    J.A. Noble. Finding corners. Image and Vision Computing, 6(2):121–128, 1988.Google Scholar
  14. 14.
    M. Okutomi and T. Kanade. A locally adaptive window for signal matching. International Journal of Computer Vision, 7(2):143–162, 1992.Google Scholar
  15. 15.
    P. Remagnino, P. Brand, and R. Mohr. Correlation techniques in adaptative template matching with uncalibrated cameras. In Vision Geometry III, SPIE's international symposium on photonic sensors & control for commercial applications, volume 2356, pages 252–253, October 1994.Google Scholar
  16. 16.
    P.J. Rousseeuw and A.M. Leroy. Robust regression and outlier detection, volume XIV of Wiley. J.Wiley and Sons, New York, 1987.Google Scholar
  17. 17.
    R. Zabih and J. Woodfill. Non-parametric local transforms for computing visual correspondance. In Proceedings of the 3rd European Conference on Computer Vision, Stockholm, Sweden, pages 151–158. Springer-Verlag, May 1994.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Zhong-Dan Lan
    • 1
  • Roger Mohr
    • 1
  • Long Quan
    • 1
  1. 1.Laboratoire Gravir, Projet MoviMontbonnotFrance

Personalised recommendations