Optimal Contour Closure by Superpixel Grouping

  • Alex Levinshtein
  • Cristian Sminchisescu
  • Sven Dickinson
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6312)


Detecting contour closure, i.e., finding a cycle of disconnected contour fragments that separates an object from its background, is an important problem in perceptual grouping. Searching the entire space of possible groupings is intractable, and previous approaches have adopted powerful perceptual grouping heuristics, such as proximity and co-curvilinearity, to manage the search. We introduce a new formulation of the problem, by transforming the problem of finding cycles of contour fragments to finding subsets of superpixels whose collective boundary has strong edge support in the image. Our cost function, a ratio of a novel learned boundary gap measure to area, promotes spatially coherent sets of superpixels. Moreover, its properties support a global optimization procedure using parametric maxflow. We evaluate our framework by comparing it to two leading contour closure approaches, and find that it yields improved performance.


Image Edge Perceptual Grouping Contour Closure Contour Image Ground Truth Segmentation 
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.
    Wertheimer, M.: Laws of organization in perceptual forms. In: Ellis, W. (ed.) Source Book of Gestalt Psychology. Harcourt, Brace (1938)Google Scholar
  2. 2.
    Estrada, F.J., Jepson, A.D.: Perceptual grouping for contour extraction. In: ICPR, pp. 32–35 (2004)Google Scholar
  3. 3.
    Estrada, F.J., Jepson, A.D.: Robust boundary detection with adaptive grouping. In: POCV, p. 184 (2006)Google Scholar
  4. 4.
    Stahl, J., Wang, S.: Edge grouping combining boundary and region information. IEEE Transactions on Image Processing 16, 2590–2606 (2007)CrossRefMathSciNetGoogle Scholar
  5. 5.
    Kolmogorov, V., Boykov, Y., Rother, C.: Applications of parametric maxflow in computer vision. In: ICCV (2007)Google Scholar
  6. 6.
    Carreira, J., Sminchisescu, C.: Constrained parametric min-cuts for automatic object segmentation. In: CVPR (2010)Google Scholar
  7. 7.
    Ren, X., Fowlkes, C.C., Malik, J.: Cue integration in figure/ground labeling. In: NIPS, pp. 1121–1128 (2005)Google Scholar
  8. 8.
    Ren, X., Fowlkes, C.C., Malik, J.: Scale-invariant contour completion using conditional random fields. In: ICCV, pp. 1214–1221 (2005)Google Scholar
  9. 9.
    Williams, L.R., Jacobs, D.W.: Stochastic completion fields: a neural model of illusory contour shape and salience. In: ICCV, p. 408 (1995)Google Scholar
  10. 10.
    Lowe, D.G.: Perceptual Organization and Visual Recognition. Kluwer Academic Publishers, Norwell (1985)Google Scholar
  11. 11.
    Brady, M., Asada, H.: Smoothed local symmetries and their implementation. IJRR 3, 36–61 (1984)Google Scholar
  12. 12.
    Cham, T.J., Cipolla, R.: Geometric saliency of curve correspondances and grouping of symmetric comntours. In: Buxton, B.F., Cipolla, R. (eds.) ECCV 1996. LNCS, vol. 1065, pp. 385–398. Springer, Heidelberg (1996)CrossRefGoogle Scholar
  13. 13.
    Saint-Marc, P., Rom, H., Medioni, G.: B-spline contour representation and symmetry detection. PAMI 15, 1191–1197 (1993)Google Scholar
  14. 14.
    Ylä-Jääski, A., Ade, F.: Grouping symmetrical structures for object segmentation and description. CVIU 63, 399–417 (1996)Google Scholar
  15. 15.
    Stahl, J.S., Wang, S.: Globally optimal grouping for symmetric closed boundaries by combining boundary and region information. PAMI 30, 395–411 (2008)Google Scholar
  16. 16.
    Levinshtein, A., Dickinson, S., Sminchisescu, C.: Multiscale Symmetric Part Detection and Grouping. In: ICCV (2009)Google Scholar
  17. 17.
    Jacobs, D.W.: Robust and efficient detection of salient convex groups. PAMI 18, 23–37 (1996)Google Scholar
  18. 18.
    Elder, J.H., Zucker, S.W.: Computing contour closure. In: Buxton, B.F., Cipolla, R. (eds.) ECCV 1996. LNCS, vol. 1065, pp. 399–412. Springer, Heidelberg (1996)CrossRefGoogle Scholar
  19. 19.
    Wang, S., Kubota, T., Siskind, J.M., Wang, J.: Salient closed boundary extraction with ratio contour. PAMI 27, 546–561 (2005)Google Scholar
  20. 20.
    Elder, J., Zucker, S.: A measure of closure. Vision Research 34, 3361–3369 (1994)CrossRefGoogle Scholar
  21. 21.
    Williams, L.R., Hanson, A.R.: Perceptual completion of occluded surfaces. CVIU 64, 1–20 (1996)Google Scholar
  22. 22.
    Zhu, Q., Song, G., Shi, J.: Untangling cycles for contour grouping. In: ICCV (2007)Google Scholar
  23. 23.
    Jermyn, I., Ishikawa, H.: Globally optimal regions and boundaries as minimum ratio weight cycles. PAMI 23, 1075–1088 (2001)Google Scholar
  24. 24.
    Maire, M., Arbelaez, P., Fowlkes, C., Malik, J.: Using contours to detect and localize junctions in natural images. In: CVPR (2008)Google Scholar
  25. 25.
    Mori, G., Ren, X., Efros, A.A., Malik, J.: Recovering human body configurations: Combining segmentation and recognition. In: CVPR, pp. 326–333 (2004)Google Scholar
  26. 26.
    Martin, D.R., Fowlkes, C.C., Malik, J.: Learning to detect natural image boundaries using local brightness, color, and texture cues. PAMI 26, 530–549 (2004)Google Scholar
  27. 27.
    Borenstein, E., Ullman, S.: Class-specific, top-down segmentation. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2351, pp. 109–124. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  28. 28.
    Alpert, S., Galun, M., Basri, R., Brandt, A.: Image segmentation by probabilistic bottom-up aggregation and cue integration. In: CVPR (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Alex Levinshtein
    • 1
  • Cristian Sminchisescu
    • 2
  • Sven Dickinson
    • 1
  1. 1.University of Toronto 
  2. 2.University of Bonn 

Personalised recommendations