Analysis of KITTI Data for Stereo Analysis with Stereo Confidence Measures

  • Ralf Haeusler
  • Reinhard Klette
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7584)


The recently published KITTI stereo dataset provides a new quality of stereo imagery with partial ground truth for benchmarking stereo matchers. Our aim is to test the value of stereo confidence measures (e.g. a left-right consistency check of disparity maps, or an analysis of the slope of a local interpolation of the cost function at the taken minimum) when applied to recorded datasets, such as published with KITTI. We choose popular measures as available in the stereo-analysis literature, and discuss a naive combination of these. Evaluations are carried out using a sparsification strategy. While the best single confidence measure proved to be the right-left consistency check for high disparity map densities, the best overall performance is achieved with the proposed naive measure combination. We argue that there is still demand for more challenging datasets and more comprehensive ground truth.


Ground Truth Stereo Matcher Disparity Estimate Disparity Error Disparity Space 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. International Journal of Computer Vision 47, 7–42 (2002)zbMATHCrossRefGoogle Scholar
  2. 2.
    The University of Auckland: .enpeda.. image sequence analysis test site (EISATS),
  3. 3.
    Heidelberg Collaboratory for Image Processing: Robust Vision Challenge,
  4. 4.
    Geiger, A., Lenz, P., Urtasun, R.: Are we ready for autonomous driving? The KITTI vision benchmark suite. In: Computer Vision and Pattern Recognition (CVPR), Providence, USA (2012)Google Scholar
  5. 5.
    Banks, J., Corke, P.I.: Quantitative evaluation of matching methods and validity measures for stereo vision. I. J. Robotic Res. 20, 512–532 (2001)CrossRefGoogle Scholar
  6. 6.
    Hu, X., Mordohai, P.: Evaluation of stereo confidence indoors and outdoors. In: [19], pp. 1466–1473Google Scholar
  7. 7.
    Wedel, A., Brox, T., Vaudrey, T., Rabe, C., Franke, U., Cremers, D.: Stereoscopic scene flow computation for 3D motion understanding. International Journal of Computer Vision 95, 29–51 (2011)zbMATHCrossRefGoogle Scholar
  8. 8.
    Shimizu, M., Okutomi, M.: Precise sub-pixel estimation on area-based matching. In: ICCV, pp. 90–97 (2001)Google Scholar
  9. 9.
    Merrell, P., Akbarzadeh, A., Wang, L., Mordohai, P., Frahm, J.-M., Yang, R., Nistér, D., Pollefeys, M.: Real-time visibility-based fusion of depth maps. In: ICCV, pp. 1–8 (2007)Google Scholar
  10. 10.
    Gehrig, S.K., Scharwächter, T.: A real-time multi-cue framework for determining optical flow confidence. In: ICCV Workshops, pp. 1978–1985 (2011)Google Scholar
  11. 11.
    Aodha, O.M., Brostow, G.J., Pollefeys, M.: Segmenting video into classes of algorithm-suitability. In: [19], pp. 1054–1061Google Scholar
  12. 12.
    Humayun, A., Aodha, O.M., Brostow, G.J.: Learning to find occlusion regions. In: CVPR, pp. 2161–2168. IEEE (2011)Google Scholar
  13. 13.
    Hirschmüller, H.: Stereo processing by semiglobal matching and mutual information. IEEE Trans. Pattern Anal. Mach. Intell. 30, 328–341 (2008)CrossRefGoogle Scholar
  14. 14.
    Egnal, G.: Mutual information as a stereo correspondence measure. Computer and Information Science, University of Pennsylvania, Philadelphia, USA, Tech. Rep. MS-CIS-00-20 (2000)Google Scholar
  15. 15.
    Hirschmüller, H., Scharstein, D.: Evaluation of cost functions for stereo matching. In: CVPR (2007)Google Scholar
  16. 16.
    Scharstein, D., Szeliski, R.: Stereo matching with nonlinear diffusion. International Journal of Computer Vision 28, 155–174 (1998)CrossRefGoogle Scholar
  17. 17.
    Shi, J., Tomasi, C.: Good features to track. In: CVPR, pp. 593–600. IEEE (1994)Google Scholar
  18. 18.
    Okutomi, M., Katayama, Y., Oka, S.: A simple stereo algorithm to recover precise object boundaries and smooth surfaces. International Journal of Computer Vision 47, 261–273 (2002)zbMATHCrossRefGoogle Scholar
  19. 19.
    The Twenty-Third IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2010, San Francisco, CA, USA, June 13-18. IEEE (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Ralf Haeusler
    • 1
  • Reinhard Klette
    • 1
  1. 1.Computer Science DepartmentThe University of AucklandNew Zealand

Personalised recommendations