Illumination Invariant Cost Functions in Semi-Global Matching

  • Simon Hermann
  • Sandino Morales
  • Tobi Vaudrey
  • Reinhard Klette
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6469)


The paper evaluates three categories of similarity measures: ordering-based (census), gradient-based, and illumination-based cost functions. The performance of those functions is evaluated especially with respect to illumination changes using two different sets of data, also including real world driving sequences of hundreds of stereo frames with strong illumination differences. The overall result is that there are cost functions in all three categories that can perform well on a quantitative and qualitative level. This leads to the assumption that those cost functions are in fact closely related at a qualitative level, and we provide our explanation.


cost functions stereo matching illumination invariance 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Aujol, J.-F., Gilboa, G., Chan, T., Osher, S.: Structure-texture image decomposition – modeling, algorithms, and parameter selection. Int. J. Computer Vision 67, 111–136 (2006)CrossRefzbMATHGoogle Scholar
  2. 2.
    Barnard, S.T., Fischler, M.A.: Computational stereo. ACM Computing Surveys 14, 553–572 (1982)CrossRefGoogle Scholar
  3. 3.
    Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. IEEE Trans. Pattern Analysis Machine Intelligence 23, 1222–1239 (2001)CrossRefGoogle Scholar
  4. 4.
    Brox, T., Bruhn, A., Papenberg, N., Weickert, J.: High accuracy optical flow estimation based on a theory for warping. In: Pajdla, T., Matas, J(G.) (eds.) ECCV 2004. LNCS, vol. 3024, pp. 25–36. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  5. 5.
    .enpeda. image sequences analysis test site,
  6. 6.
    El-Mahassani, E.D.: New robust matching cost functions for stereo vision. In: Proc. DICTA, pp. 144–150 (2007)Google Scholar
  7. 7.
    Felzenszwalb, P.F., Huttenlocher, D.: Efficient belief propagation for early vision. Int. J. Computer Vision 70, 41–54 (2006)CrossRefGoogle Scholar
  8. 8.
    Gehrig, S.K., Eberli, F., Meyer, T.: A real-time low-power stereo vision engine using semi-global matching. In: Proc. ICCV, pp. 134–143 (2009)Google Scholar
  9. 9.
    Haeusler, R., Klette, R.: Benchmarking stereo data (Not the matching algorithms). In: Goesele, M., Roth, S., Kuijper, A., Schiele, B., Schindler, K. (eds.) Pattern Recognition. LNCS, vol. 6376, pp. 383–392. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  10. 10.
    Hirschmüller, H.: Accurate and efficient stereo processing by semi-global matching and mutual information. In: Proc. CVPR, vol. 2, pp. 807–814 (2005)Google Scholar
  11. 11.
    Hirschmüller, H., Scharstein, D.: Evaluation of cost functions for stereo matching. In: Proc. CVPR, pp. 1–8 (2007)Google Scholar
  12. 12.
    Hirschmüller, H., Scharstein, D.: Evaluation of stereo matching costs on images with radiometric differences. IEEE Trans. Pattern Analysis Machine Intelligence 31, 1582–1599 (2009)CrossRefGoogle Scholar
  13. 13.
    Klaus, A., Sormann, M., Karner, K.: Segment-based stereo matching using belief propagation and a self-adapting dissimilarity measure. In: Proc. CVPR, vol. 3, pp. 15–18 (2006)Google Scholar
  14. 14.
    Middlebury College, stereo vision page,
  15. 15.
    Morales, S., Vaudrey, T., Klette, R.: A third eye for performance evaluation in stereo sequence analysis. In: Jiang, X., Petkov, N. (eds.) CAIP 2009. LNCS, vol. 5702, pp. 1078–1086. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  16. 16.
    Morales, S., Woo, Y.W., Klette, R., Vaudrey, T.: A study on stereo and motion data accuracy for a moving platform. In: Kim, J.-H., Ge, S.S., Vadakkepat, P., Jesse, N., Al Manum, A., Puthusserypady, S.K., Rückert, U., Sitte, J., Witkowski, U., Nakatsu, R., Braunl, T., Baltes, J., Anderson, J., Wong, C.-C., Verner, I., Ahlgren, D. (eds.) Advances in Robotics. LNCS, vol. 5744, pp. 292–300. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  17. 17.
    Ohta, Y., Kanade, T.: Stereo by two-level dynamic programming. In: Proc. Int. Joint Conf. Artificial Intelligence, vol. 2, pp. 1120–1126 (1985)Google Scholar
  18. 18.
    Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. In: Proc. ICCV, pp. 7–42 (2002)Google Scholar
  19. 19.
    Szeliski, R.: Prediction error as a quality metric for motion and stereo. In: Proc. ICCV, pp. 781–788 (1999)Google Scholar
  20. 20.
    Vaudrey, T., Klette, R.: Residual images remove illumination artifacts! In: Denzler, J., Notni, G., Süße, H. (eds.) Pattern Recognition. LNCS, vol. 5748, pp. 472–481. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  21. 21.
    Vaudrey, T., Wedel, A., Klette, R.: A methodology for evaluating illumination artifact removal for corresponding images. In: Jiang, X., Petkov, N. (eds.) CAIP 2009. LNCS, vol. 5702, pp. 1113–1121. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  22. 22.
    Zabih, R., Woodfill, J.: Non-parametric local transform for computing visual correspondence. In: Proc. ECCV, vol. 2, pp. 151–158 (1994)Google Scholar
  23. 23.
    Zach, C., Pock, T., Bischof, H.: A duality based approach for realtime TV-L1 optical flow. In: Hamprecht, F.A., Schnörr, C., Jähne, B. (eds.) DAGM 2007. LNCS, vol. 4713, pp. 214–223. Springer, Heidelberg (2007)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Simon Hermann
    • 1
  • Sandino Morales
    • 1
  • Tobi Vaudrey
    • 1
  • Reinhard Klette
    • 1
  1. 1.enpeda.. group, Dept. Computer ScienceUniversity of AucklandNew Zealand

Personalised recommendations