An important function of perceptual grouping is the restoration of contours. Edge maps produced by low level edge detectors are invariably noisy and inconsistent. It it the aim of perceptual grouping to refine these edge segments by imposing consistency based on considerations about real object outlines. In this paper we describe a method for grouping edge segments into perceptually salient contours using splines. The two important ingredients of our method are firstly the use of probability distributions for possible orientation structure in the image, and secondly the use of Kellman-Shipley relatability to find perceptually meaningful structure. The spline parameters are adjusted to optimise their probabilities in terms of image structure and bending. Consistent structure is then identified using both perceptual criteria and similarity to contour structure in the image.


Spline Interpolation Perceptual Organisation Perceptual Grouping Contour Segment Edge Segment 
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.


  1. 1.
    Sha’ashua, A., Ullman, S.: Structural saliency: The detection of globally salient structures using a locally connected network. In: Proceedings of the 2nd ICCV, pp. 321–327 (1988); Also Weizmann Institute of Science Report CS88-18 (October 1988)Google Scholar
  2. 2.
    Sarkar, S., Boyer, K.: Computing Perceptual Organisation in Computer Vision. World Scientific, Singapore (1994)CrossRefGoogle Scholar
  3. 3.
    Elder, J., Zucker, S.: Computing contour closure. In: Buxton, B.F., Cipolla, R. (eds.) ECCV 1996. LNCS, vol. 1064, p. 399. Springer, Heidelberg (1996)CrossRefGoogle Scholar
  4. 4.
    Guy, G., Medioni, G.: Inferring global perceptual contours from local features. Int. J. Comp. Vis. 20, 113–133 (1996)CrossRefGoogle Scholar
  5. 5.
    Kellman, P., Shipley, T.: A theory of visual interpolation in object perception. Cognitive Psychology 23, 141–221 (1991)CrossRefGoogle Scholar
  6. 6.
    Lüdtke, N., Wilson, R., Hancock, E.: Population codes for orientation estimation. In: Proceedings of the 15th International Conference on Pattern Recognition (2000)Google Scholar
  7. 7.
    Lüdtke, N., Wilson, R., Hancock, E.: Probabilistic population coding of multiple edge orientation. In: Proceedings of the International Conference on Image Processing, IEEE, Los Alamitos (2002) (to appear)Google Scholar
  8. 8.
    Parent, P., Zucker, S.: Trace inference, curvature consistency, and curve detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 11, 823–839 (1989)CrossRefGoogle Scholar
  9. 9.
    Hancock, E., Kittler, J.: Discrete relaxation. Pattern Recognition 23, 711–733 (1990)CrossRefGoogle Scholar
  10. 10.
    Gavrila, D.: Hermite deformable contours. Technical Report CS-TR-3610, Center for Automation Research, University of Maryland (1996); Also ICPR 1996Google Scholar
  11. 11.
    Iverson, L., Zucker, S.: Logical/linear operators for image curves. IEEE Transactions on Pattern Recognition and Machine Intelligence 17, 982–996 (1995)CrossRefGoogle Scholar
  12. 12.
    Koendrink, J., van Doorn, A.: The shape of smooth objects and the way contours end. Perception 11, 129–137 (1982)CrossRefGoogle Scholar
  13. 13.
    Canny, J.: A computational approach to edge detection. IEEE Transact. on Patt. Rec. and Machine Intell. 8, 679–700 (1986)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Niklas Ludtke
    • 1
    • 2
  • Richard C. Wilson
    • 1
    • 2
  1. 1.Beckman InstituteUniversity of IllinoisUSA
  2. 2.Dept. of Computer ScienceUniversity of YorkUK

Personalised recommendations