Robust Fitting by Adaptive-Scale Residual Consensus

  • Hanzi Wang
  • David Suter
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3023)


Computer vision tasks often require the robust fit of a model to some data. In a robust fit, two major steps should be taken: i) robustly estimate the parameters of a model, and ii) differentiate inliers from outliers. We propose a new estimator called Adaptive-Scale Residual Consensus (ASRC). ASRC scores a model based on both the residuals of inliers and the corresponding scale estimate determined by those inliers. ASRC is very robust to multiple-structural data containing a high percentage of outliers. Compared with RANSAC, ASRC requires no pre-determined inlier threshold as it can simultaneously estimate the parameters of a model and the scale of inliers belonging to that model. Experiments show that ASRC has better robustness to heavily corrupted data than other robust methods. Our experiments address two important computer vision tasks: range image segmentation and fundamental matrix calculation. However, the range of potential applications is much broader than these.


Standard Variance Fundamental Matrix Robust Estimator Point Pair Breakdown Point 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Bab-Hadiashar, A., Suter, D.: Robust Optic Flow Computation. International Journal of Computer Vision 29(1), 59–77 (1998)CrossRefGoogle Scholar
  2. 2.
    Chen, H., Meer, P.: Robust Regression with Projection Based M-estimators. In: ICCV, Nice, France, pp. 878–885 (2003)Google Scholar
  3. 3.
    Comaniciu, D., Meer, P.: Mean Shift Analysis and Applications. In: ICCV, Kerkyra, Greece, pp. 1197–1203 (1999)Google Scholar
  4. 4.
    Comaniciu, D., Ramesh, V., Bue, A.D.: Multivariate Saddle Point Detection for Statistical Clustering. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2352, pp. 561–576. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  5. 5.
    Fischler, M.A., Rolles, R.C.: Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Commun. ACM. 24(6), 381–395 (1981)CrossRefGoogle Scholar
  6. 6.
    Haralick, R.M.: Computer vision theory: The lack thereof. CVGIP 36, 372–386 (1986)Google Scholar
  7. 7.
    Hoover, A., Jean-Baptiste, G., Jiang, X.: An Experimental Comparison of Range Image Segmentation Algorithms. IEEE Trans. PAMI. 18(7), 673–689 (1996)Google Scholar
  8. 8.
    Hough, P.V.C.: Methods and means for recognising complex patterns. U.S. Patent 3 069 654 (1962)Google Scholar
  9. 9.
    Huber, P.J.: Robust Statistics, New York. Wiley, Chichester (1981)zbMATHCrossRefGoogle Scholar
  10. 10.
    Lee, K.-M., Meer, P., Park, R.-H.: Robust Adaptive Segmentation of Range Images. IEEE Trans. PAMI. 20(2), 200–205 (1998)Google Scholar
  11. 11.
    Miller, J.V., Stewart, C.V.: MUSE: Robust Surface Fitting Using Unbiased Scale Estimates. In: CVPR, San Francisco, pp. 300–306 (1996)Google Scholar
  12. 12.
    Rousseeuw, P.J., Croux, C.: Alternatives to the Median Absolute Derivation. Journal of the American Statistical Association 88(424), 1273–1283 (1993)zbMATHCrossRefMathSciNetGoogle Scholar
  13. 13.
    Rousseeuw, P.J., Leroy, A.: Robust Regression and outlier detection. John Wiley & Sons, New York (1987)zbMATHCrossRefGoogle Scholar
  14. 14.
    Silverman, B.W.: Density Estimation for Statistics and Data Analysis London. Chapman and Hall, Boca Raton (1986)Google Scholar
  15. 15.
    Stewart, C.V.: MINPRAN: A New Robust Estimator for Computer Vision. IEEE Trans. PAMI. 17(10), 925–938 (1995)Google Scholar
  16. 16.
    Stewart, C.V.: Bias in Robust Estimation Caused by Discontinuities and Multiple Structures. IEEE Trans. PAMI 19(8), 818–833 (1997)Google Scholar
  17. 17.
    Torr, P., Murray, D.: The Development and Comparison of Robust Methods for Estimating the Fundamental Matrix. International Journal of Computer Vision 24, 271–300 (1997)CrossRefGoogle Scholar
  18. 18.
    Torr, P., Zisserman, A.: MLESAC: A New Robust Estimator With Application to Estimating Image Geometry. Computer Vision and Image Understanding 78(1), 138–156 (2000)CrossRefGoogle Scholar
  19. 19.
    Wand, M.P., Jones, M.: Kernel Smoothing. Chapman & Hall, Boca Raton (1995)zbMATHGoogle Scholar
  20. 20.
    Wang, H., Suter, D.: False-Peaks-Avoiding Mean Shift Method for Unsupervised Peak- Valley Sliding Image Segmentation. In: Digital Image Computing Techniques and Applications, Sydney, Australia, pp. 581–590 (2003)Google Scholar
  21. 21.
    Wang, H., Suter, D.: MDPE: A Very Robust Estimator for Model Fitting and Range Image Segmentation. International Journal of Computer Vision (2003) (to appear)Google Scholar
  22. 22.
    Wang, H., Suter, D.: Variable bandwidth QMDPE and its application in robust optic flow estimation. In: ICCV, Nice, France, pp. 178–183 (2003)Google Scholar
  23. 23.
    Wang, H., Suter, D.: Robust Adaptive-Scale Parametric Model Estimation for Computer Vision. Submitted to IEEE Trans. PAMI (2003)Google Scholar
  24. 24.
    Yu, X., Bui, T.D., Krzyzak, A.: Robust Estimation for Range Image Segmentation and Reconstruction. IEEE Trans PAMI. 16(5), 530–538 (1994)Google Scholar
  25. 25.
    Zhang, Z., et al.: A Robust Technique for Matching Two Uncalibrated Image Through the Recovery of the Unknown Epipolar Geometry. Artificial Intelligence 78, 11–87 (1995)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Hanzi Wang
    • 1
  • David Suter
    • 1
  1. 1.Department of Electrical and Computer Systems EngineeringMonash UniversityClaytonAustralia

Personalised recommendations