Improving Depth Map Quality with Markov Random Fields

  • Rafał Kozik
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 102)


Currently there is an increasing number of solutions adapting sterevision camera for depth perception. Thanks to the two slightly different projections of the same scene it is possible to estimate distance to particular object. However the commononly used real-time correlation-based solutions usually suffer from inaccuracy caused by low textured regions or occlusions. Therefore in this article an statistical model-based approach for depth estimation is proposed. It engages both stereovision camera and prior knowledge of scene structure.


Feature Vector Depth Estimation Markov Random Depth Discontinuity Markov Random Field Model 
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.
    Saxena, A., Chung, S.H., Ng, A.Y.: Learning Depth from Single Monocular Images. Neural Information Processing Systems (NIPS) 18 (2005)Google Scholar
  2. 2.
    Needleman, S.B., Wunsch, C.D.: A general method applicable to the search for similarities in the amino acid sequence of two proteins. Journal of Molecular Biology (1970)Google Scholar
  3. 3.
    Birchfield, S., Tomasi, C.: Depth Discontinuities by Pixel-to-Pixel Stereo. International Journal of Computer Vision 35(3), 269–293 (1999)CrossRefGoogle Scholar
  4. 4.
  5. 5.
    Daniel, P., Huttenlocher, P., Felzenszwalb, B.: Efficient Belief Propagation for Early Vision. Efficient Belief Propagation for Early Vision (2006)Google Scholar
  6. 6.
    Pelcztnski, P.: Travel Aid System for the Blind. Image Processing and Communications Challenges, 324–333 (2009)Google Scholar
  7. 7.
    The Miniguide project homepage,
  8. 8.
    Sun, J., Li, Y., Kang, S., Shum, H.: Symmetric stereo matching for occlusion handling. In: CVPR, pp. II: 399–II: 406 (2005)Google Scholar
  9. 9.
    Kolmogorov, V., Zabih, R.: Computing visual correspondence with occlusions via graph cuts, pp. II: 508–II: 515 (2001)Google Scholar
  10. 10.
    Sivic, J., Zisserman, A.: Efficient Visual Search for Objects in Videos. Proceedings of the IEEE (2008)Google Scholar
  11. 11.
    Kirasic, D., Basch, D.: Ontology-Based Design Pattern Recognition. LNCS (2009)Google Scholar
  12. 12.
    Francois, A.R.J., Nevatia, R., Hobbs, J., Bolles, R.C.: VERL: an ontology framework for representing and annotating video events. IEEE MultiMedia 12(4), 76–86 (2005)CrossRefGoogle Scholar
  13. 13.
    Latfi, F., Lefebvre, B., Descheneaux, C.: Ontology-based management of the telehealth smart home, dedicated to elderly in loss of cognitive autonomy. In: CEUR Workshop Proceedings, vol. 258 (2007)Google Scholar
  14. 14.
    Torralba, R., Fergus, Weiss, Y.: Small codes and large databases for recognition. In CVPR (2008)Google Scholar
  15. 15.
    Philbin, J., Chum, O., Isard, M., Sivic, J., Zisserman, A.: Lost in quantization: Improving particular object retrieval in large scale image databases. In: CVPR (2008)Google Scholar
  16. 16.
    Felzenszwalb, P.F., Huttenlocher, D.P.: Efficient belief propagation for early vision. In: CVPR, pp. 1:261–1:268 (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Rafał Kozik
    • 1
  1. 1.Institute of TelecommunicationsUniversity of Technology & Life SciencesBydgoszczPoland

Personalised recommendations