Local Occlusion Detection under Deformations Using Topological Invariants

  • Edgar Lobaton
  • Ram Vasudevan
  • Ruzena Bajcsy
  • Ron Alterovitz
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6313)


Occlusions provide critical cues about the 3D structure of man-made and natural scenes. We present a mathematical framework and algorithm to detect and localize occlusions in image sequences of scenes that include deforming objects. Our occlusion detector works under far weaker assumptions than other detectors. We prove that occlusions in deforming scenes occur when certain well-defined local topological invariants are not preserved. Our framework employs these invariants to detect occlusions with a zero false positive rate under assumptions of bounded deformations and color variation. The novelty and strength of this methodology is that it does not rely on spatio-temporal derivatives or matching, which can be problematic in scenes including deforming objects, but is instead based on a mathematical representation of the underlying cause of occlusions in a deforming 3D scene. We demonstrate the effectiveness of the occlusion detector using image sequences of natural scenes, including deforming cloth and hand motions.


Image Sequence Motion Estimate Color Variation IEEE Conf Topological Invariant 
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.

Supplementary material

978-3-642-15558-1_8_MOESM1_ESM.gif (187 kb)
Electronic Supplementary Material (188 KB)
978-3-642-15558-1_8_MOESM2_ESM.gif (187 kb)
Electronic Supplementary Material (188 KB)
978-3-642-15558-1_8_MOESM3_ESM.gif (1.4 mb)
Electronic Supplementary Material (1,480 KB)
978-3-642-15558-1_8_MOESM4_ESM.gif (1.4 mb)
Electronic Supplementary Material (1,480 KB)
978-3-642-15558-1_8_MOESM5_ESM.gif (1.7 mb)
Electronic Supplementary Material (1,691 KB)
978-3-642-15558-1_8_MOESM6_ESM.gif (1.7 mb)
Electronic Supplementary Material (1,691 KB)


  1. 1.
    Apostoloff, N., Fitzgibbon, A.: Learning Spatiotemporal T-Junctions for Occlusion Detection. In: IEEE Conf. on Computer Vision and Pattern Recognition (2005)Google Scholar
  2. 2.
    Apostoloff, N., Fitzgibbon, A.: Automatic video segmentation using spatiotemporal T-junctions. In: British Machine Vision Conf., vol. 3, p. 1089 (2006)Google Scholar
  3. 3.
    Favaro, P., Duci, A., Ma, Y., Soatto, S.: On exploiting occlusions in multiple-view geometry. In: IEEE Intl. Conf. on Computer Vision, pp. 479–486 (2003)Google Scholar
  4. 4.
    Irani, M., Rousso, B., Peleg, S.: Computing occluding and transparent motions. Intl. J. of Computer Vision 12, 5–16 (1994)CrossRefGoogle Scholar
  5. 5.
    Niyogi, S., Adelson, E.: Analyzing and recognizing walking figures in XYT. In: IEEE Conf. on Computer Vision and Pattern Recognition, pp. 469–474 (1994)Google Scholar
  6. 6.
    Stein, A., Hebert, M.: Local detection of occlusion boundaries in video. Image and Vision Computing (2008)Google Scholar
  7. 7.
    Horn, B., Schunck, B.: Determining Optical Flow. Artificial Intelligence 17, 185–203 (1981)CrossRefGoogle Scholar
  8. 8.
    Biederman, I.: Recognition-by-components: A theory of human image understanding. Psychological Review 94, 115–147 (1987)CrossRefGoogle Scholar
  9. 9.
    Ogale, A., Fermuller, C., Aloimonos, Y.: Motion segmentation using occlusions. IEEE Trans. on Pattern Analysis and Machine Intelligence 27, 988–992 (2005)CrossRefGoogle Scholar
  10. 10.
    Smith, P., Drummond, T., Cipolla, R.: Layered motion segmentation and depth ordering by tracking edges. IEEE Trans. on Pattern Analysis and Machine Intelligence 26, 479–494 (2004)CrossRefGoogle Scholar
  11. 11.
    Xiao, J., Shah, M.: Accurate motion layer segmentation and matting. In: IEEE Conf. on Computer Vision and Pattern Recognition, vol. 2 (2005)Google Scholar
  12. 12.
    Yin, P., Criminisi, A., Winn, J., Essa, I.: Tree-based classifiers for bilayer video segmentation. In: IEEE Conf. on Computer Vision and Pattern Recognition (2007)Google Scholar
  13. 13.
    Anandan, P.: A computational framework and an algorithm for the measurement of visual motion. Intl. J. of Computer Vision 2, 283–310 (1989)CrossRefGoogle Scholar
  14. 14.
    Black, M., Anandan, P.: Robust dynamic motion estimation over time. In: IEEE Conf. on Computer Vision and Pattern Recognition, pp. 296–302 (1991)Google Scholar
  15. 15.
    Black, M., Fleet, D.: Probabilistic detection and tracking of motion discontinuities. Intl. J. of Computer Vision 38, 231–245 (2000)zbMATHCrossRefGoogle Scholar
  16. 16.
    Fleet, D., Black, M., Nestares, O.: Bayesian inference of visual motion boundaries. Morgan Kaufmann Publishers Inc., San Francisco (2003)Google Scholar
  17. 17.
    Nestares, O., Fleet, D.: Probabilistic tracking of motion boundaries with spatiotemporal predictions. In: IEEE Conf. on Computer Vision and and Pattern Recognition, vol. 2, pp. 358–365 (2001)Google Scholar
  18. 18.
    Stein, A., Hoiem, D., Hebert, M.: Learning to Find Object Boundaries Using Motion Cues. In: IEEE Intl. Conf. on Computer Vision, pp. 1–8 (2007)Google Scholar
  19. 19.
    Beymer, D.: Finding Junctions Using the Image Gradient. Massachusetts Institute of Technology, Artificial Intelligence Laboratory (1991)Google Scholar
  20. 20.
    Freeman, W., Adelson, E.: The design and use of steerable filters. IEEE Trans. on Pattern Analysis and Machine Intelligence 13, 891–906 (1991)CrossRefGoogle Scholar
  21. 21.
    Li, D., Sullivan, G., Baker, K.: Edge detection at junctions. In: Alvey Vision Conf., vol. 2 (1989)Google Scholar
  22. 22.
    Perona, P.: Steerable-Scalable Kernels for Edge Detection and Junction Analysis. In: Sandini, G. (ed.) ECCV 1992. LNCS, vol. 588, pp. 3–18. Springer, Heidelberg (1992)Google Scholar
  23. 23.
    Parida, L., Geiger, D., Hummel, R.: Junctions: detection, classification, and reconstruction. IEEE Trans. on Pattern Analysis and Machine Intelligence 20, 687–698 (1998)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Edgar Lobaton
    • 1
  • Ram Vasudevan
    • 2
  • Ruzena Bajcsy
    • 2
  • Ron Alterovitz
    • 1
  1. 1.Department of Computer ScienceUniversity of North Carolina at Chapel Hill
  2. 2.Department of Electrical Engineering and Computer SciencesUniversity of CaliforniaBerkeley

Personalised recommendations