Learning Class-Specific Edges for Object Detection and Segmentation

  • Mukta Prasad
  • Andrew Zisserman
  • Andrew Fitzgibbon
  • M. Pawan Kumar
  • P. H. S. Torr
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4338)


Recent research into recognizing object classes (such as humans, cows and hands) has made use of edge features to hypothesize and localize class instances. However, for the most part, these edge-based methods operate solely on the geometric shape of edges, treating them equally and ignoring the fact that for certain object classes, the appearance of the object on the “inside” of the edge may provide valuable recognition cues.

We show how, for such object classes, small regions around edges can be used to classify the edge into object or non-object. This classifier may then be used to prune edges which are not relevant to the object class, and thereby improve the performance of subsequent processing. We demonstrate learning class specific edges for a number of object classes — oranges, bananas and bottles — under challenging scale and illumination variation.

Because class-specific edge classification provides a low-level analysis of the image it may be integrated into any edge-based recognition strategy without significant change in the high-level algorithms. We illustrate its application to two algorithms: (i) chamfer matching for object detection, and (ii) modulating contrast terms in MRF based object-specific segmentation. We show that performance of both algorithms (matching and segmentation) is considerably improved by the class-specific edge labelling.


Support Vector Machine Gaussian Mixture Model Object Detection Boundary Term Object Class 
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.
    Belongie, S., Malik, J., Puzicha, J.: Matching shapes. In: Proc. ICCV, vol. 1, pp. 454–461 (2001)Google Scholar
  2. 2.
    Boykov, Y., Jolly, M.: Interactive graph cuts for optimal boundary and region segmentation of objects in n-d images. In: ICCV, pp. 105–112 (2001)Google Scholar
  3. 3.
    Carmichael, O., Mahamud, S., Hebert, M.: Discriminant filters for object recognition. Technical Report CMU-RI-TR-02-09, Carnegie Mellon University (March 2002)Google Scholar
  4. 4.
    Dollar, P., Zhuowen, T., Belongie, S.: Supervised learning of edges and object boundaries. In: Proc. CVPR (2006)Google Scholar
  5. 5.
    Ferrari, V., Tuytelaars, T., Van Gool, L.J.: Object detection by contour segment networks. pp. 14–28 (2006)Google Scholar
  6. 6.
    Gavrila, D.: Pedestrian detection from a moving vehicle. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1843, pp. 37–49. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  7. 7.
    Huttenlocher, D.P., Ullman, S.: Object recognition using alignment. In: Proc. ICCV, pp. 102–111 (1987)Google Scholar
  8. 8.
    Joachims, T.: Making large-scale support vector machine learning practical. In: Smola, A., Schölkopf, B., Burges, C. (eds.) Advances in Kernel Methods: Support Vector Machines, MIT Press, Cambridge (1998)Google Scholar
  9. 9.
    Kumar, M.P., Torr, P.H.S., Zisserman, A.: Extending pictorial structures for object recognition. In: Proc. BMVC (2004)Google Scholar
  10. 10.
    Kumar, M.P., Torr, P.H.S., Zisserman, A.: OBJ CUT. In: Proc. CVPR, pp. 18–25 (2005)Google Scholar
  11. 11.
    Lowe, D.G.: Three-dimensional object recognition from single two-dimensional images. Artificial Intelligence 31(3), 355–395 (1987)CrossRefGoogle Scholar
  12. 12.
    McHenry, K., Ponce, J.: A geodesic active contour framework for finding glass. In: Proc. CVPR (2006)Google Scholar
  13. 13.
    McHenry, K., Ponce, J., Forsyth, D.A.: Finding glass. In: Proc. CVPR, pp. 973–979 (2005)Google Scholar
  14. 14.
    Opelt, A., Pinz, A., Zisserman, A.: Incremental learning of object detectors using a visual alphabet. In: Proc. CVPR (2006)Google Scholar
  15. 15.
    Rothwell, C., Zisserman, A., Forsyth, D., Mundy, J.: Canonical frames for planar object recognition. In: Sandini, G. (ed.) ECCV 1992. LNCS, vol. 588, Springer, Heidelberg (1992)Google Scholar
  16. 16.
    Scholkopf, B., Smola, A.: Learning with Kernels. MIT Press, Cambridge (2002)Google Scholar
  17. 17.
    Seemann, E., Leibe, B., Mikolajczyk, K., Schiele, B.: An evaluation of local shape-based features for pedestrian detection. In: Proc. BMVC (2005)Google Scholar
  18. 18.
    Shahrokni, A., Fleuret, F., Fua, P.: Classifier-based contour tracking for rigid and deformable objects. In: Proc. BMVC, Oxford, UK (2005)Google Scholar
  19. 19.
    Shotton, J., Blake, A., Cipolla, R.: Contour-based learning for object detection. In: Proc. ICCV, pp. 503–510 (2005)Google Scholar
  20. 20.
    Sidenbladh, H., Black, M.: Learning image statistics for bayesian tracking. In: Proc. ECCV (2001)Google Scholar
  21. 21.
    Stenger, B., Thayananthan, A., Torr, P.H.S., Cipolla, R.: Hand pose estimation using hierarchical detection. In: CVHCI 2004, pp. 105–116 (2004)Google Scholar
  22. 22.
    Sun, J., Zhang, W., Tang, X., Shum, H.: Background cut. In: Proc. ECCV (2006)Google Scholar
  23. 23.
    Thayananthan, A., Stenger, B., Torr, P.H.S., Cipolla, R.: Shape context and chamfer matching in cluttered scenes. In: Proc. CVPR, pp. 127–133 (2003)Google Scholar
  24. 24.
    Tu, Z.: Probabilistic boosting-tree: Learning discriminative models for classification, recognition, and clustering. In: Proc. ICCV, pp. 1589–1596 (2005)Google Scholar
  25. 25.
    Varma, M., Zisserman, A.: Texture classification: Are filter banks necessary? In: Proc. CVPR, June 2003, vol. 2, pp. 691–698 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Mukta Prasad
    • 1
  • Andrew Zisserman
    • 1
  • Andrew Fitzgibbon
    • 2
  • M. Pawan Kumar
    • 3
  • P. H. S. Torr
    • 3
  1. 1.University of OxfordU.K.
  2. 2.Microsoft ResearchCambridgeU.K.
  3. 3.Oxford Brookes UniversityU.K.

Personalised recommendations