International Journal of Computer Vision

, Volume 14, Issue 2, pp 119–130 | Cite as

Context-free attentional operators: The generalized symmetry transform

  • Daniel Reisfeld
  • Haim Wolfson
  • Yehezkel Yeshurun


Active vision systems, and especially foveated vision systems, depend on efficient attentional mechanisms. We propose that machine visual attention should consist of both high-level, context-dependent components, and low-level, context free components. As a basis for the context-free component, we present an attention operator based on the intuitive notion of symmetry, which generalized many of the existing methods of detecting regions of interest. It is a low-level operator that can be applied successfully without a priori knowledge of the world. The resultingsymmetry edge map can be applied in various low, intermediate-and high- level tasks, such as extraction of interest points, grouping, and object recognition. In particular, we have implemented an algorithm that locates interest points in real time, and can be incorporated in active and purposive vision systems. The results agree with some psychophysical findings concerning symmetry as well as evidence concerning selection of fixation points. We demonstrate the performance of the transform on natural, cluttered images.


Computer Vision Vision System Computer Image Object Recognition Visual Attention 
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. Aboot, A.L. and Ahuja, N., 1988. Surface reconstruction by dynamic integration of focus, camera vergence and stereo,Proc. 2nd Intern. Conf. Comput. Vis., Tampa, FL.Google Scholar
  2. Aloimonos, J.Y., Weiss, I., and Bandyopadhyay, A., 1987. Active vision,Intern. J. Comput. Vis., 1:334–356.Google Scholar
  3. Antes, J.R., 1974. The time course of picture viewing,J. Psychol., 103:62–70.Google Scholar
  4. Atallah, M.J., 1985. On symmetry detection,IEEE Trans Comput. C-34:663–666.Google Scholar
  5. Attneave, F., 1954. Informational aspects of visual perception,Psychological Review, 61:183–193.Google Scholar
  6. Bajcsy, R., 1988. Active perception.Proc. IEEE, 76(8):996–1006.Google Scholar
  7. Ballard, D., 1990. Animated vision, Tech. Rep. TR 61, University of Rochester, Department of Computer Science, 1990.Google Scholar
  8. Bigun, J., 1988. Pattern recognition by detection of local symmetries. In E.S. Gelsma and L.N. Kanal, eds., Pattern Recognition and Artificial Intelligence, Elsevier, North Holland, pp. 75–90.Google Scholar
  9. Blum, H. and Nagel, R.N., 1978. Shape description using weighted symmetric axis features,Pattern Recognition, 10:167–180.Google Scholar
  10. Bonneh, Y., Reisfeld, D., and Yeshurun, Y., 1993. Texture discrimination by local generalized symmetry,Proc. 4th Intern. Conf. Comput. Vis., Berlin.Google Scholar
  11. Brady, M. and Asada, H., 1984. Smooth local symmetries and their implementation,Intern. J. Robot. Res. 3(3):36–61.Google Scholar
  12. Brunnstrome, K., Lindeberg, T., and Eklundh, J.O., 1992. Active detection and classification of junctions by foveation with a head-eye system guided by the scale-space primal sketch,Proc. 2nd Europ. Conf. Comput. Vis., Santa Margherita, Ligure, Italy, 7:701–709.Google Scholar
  13. Cohen, K., 1981. The development of strategies of of visual search, in eye movements. In D. Fisher, R. Monty, and J. Senders, Ed., Cognition and Visual Perception, Laurence Erlbaum Assoc: Hillsdale NJ, pp. 299–314.Google Scholar
  14. Crowley, J.L., 1991. Towards continuously operating integrated vision systems for robotics applications,SCIA-91, 7th Scandinavian Conf. Image Anal, Aalborg.Google Scholar
  15. Culhane, S.M. and Tsotsos, J.K., 1992. An attentional prototype for early vision,Proc. 2nd Europ. Conf. Comput. Vis., S. Margherita, Ligure, Italy, May, pp. 551–560.Google Scholar
  16. Davis, L.S., 1977. Understanding shape: Ii. symmetry,IEEE Trans. Syst. Man, Cybern. 7:204–211.Google Scholar
  17. Edelman, S., Reisfeld, D., and Yeshurun, Y., 1992. Learning to recognize faces from examples,Proc. 2nd Europ. Conf. Comput. Vis., Santa Margherita, Ligure, Italy, 7:787–791.Google Scholar
  18. Haith, M.M., Bergman, T., and Moore, M.J., 1977, Eye contact and face scanning in early infancy, Science 198:853–855.Google Scholar
  19. Kanade, J. and Kender, J.P., 1983. Mapping image properties into shape constraints: skewed symmetry, affine-transformable patterns, and the shape-from-texture paradigm. In Beck, Hope and Rosenfeld, eds., Human and Machine Vision, Academic Press: New York.Google Scholar
  20. Kaufman, L. and Richards, W. 1969. Spontaneous fixation tendencies for visual forms,Perception and Psychophysics, 5(2):85–88.Google Scholar
  21. Lamdan, Y., Schwartz, J.T., and Wolfson, H., 1988. On recognition of 3-d objects from 2-d images,Proc. IEEE Intern. Conf. Robot. Autom. Philadelphia, 1407–1413.Google Scholar
  22. Locher, P.J. and Nodine, C.F., 1986. Symmetry catches the eye. In A. Levy Schoeh (ed.),Eye Movements: From Physiology to Cognition, Elsevier: North Holland, pp. 353–361.Google Scholar
  23. Loftus, G. and Mackworth, N., 1978. Cognitive determinants of fixation location during picture viewing,Human Perception and Performance 4:565–572.Google Scholar
  24. Marola, G., 1989. On the detection of the axis of symmetry of symmetric and almost symmetric plannar images,IEEE Trans. Patt. Anal. Mach. Intell., 11(1):104–108.Google Scholar
  25. Moravec, H.P., 1977. Towards automatic visual obstacle avoidance,5th Intern. Joint Conf. Artif. Intell. Cambridge, MA, pp. 584–590.Google Scholar
  26. Nevatia, R. and Binford, T.O. 1977. Description and recognition of curved objects.Artificial Intelligence, 8:77–98.Google Scholar
  27. Posner, M.L. and Peterson, S.E., 1990. The attention system of the human brain.,Annu. Rev. Neurosci. 13:25–42.Google Scholar
  28. Reisfeld, D., 1994.Generalized symmetry transforms: attentional mechanisms and face recognition, Ph.D. thesis, Computer Science Department, Tel-Aviv University, January.Google Scholar
  29. Reisfeld, D., Wolfson, H., and Yeshurun, Y., 1990. Detection of interest points using symmetry,Proc. 3rd Intern. Conf. Comput. Vis., Osaka, Japan, December, pp. 62–65.Google Scholar
  30. Reisfeld, D. and Yeshurun, Y., 1992. Robust detection of facial features by generalized symmetry,Proc. 11th Intern. Conf. Image Anal. Patt. Recog., The Hague, The Netherlands, August, pp. 117–120.Google Scholar
  31. Rimey, R.D. and Brown, C.M., 1992. Where to look next using a bayes net: incorporating geometric relations,Proc. 2nd Europ. Conf. Comput. Vis., S. Margherita, Ligure, Italy, May, pp. 542–550.Google Scholar
  32. Rojer, A. and Schwartz, E., 1990. Design considerations for a space-variant visual sensor with complex logarithmic geometry,Proc. 10th Intern. Conf. Patt. Recog., pp. 278–285.Google Scholar
  33. Salapatek, P. and Kessen, W., 1973. Prolonged investigation of a plane geometric triangle by the human newborn,J. Exper. Child Psychol. 15:22–29.Google Scholar
  34. Tistarelli, M. and Sandini, G., 1990. Estimation of depth from motion using an anthropomorphic visual sensor,Image Vis. Comput., 8(4):271–278.Google Scholar
  35. Ullman, S., 1984. Visual routines,Cognition, 18:97–159.Google Scholar
  36. Xia, Y., 1989. Skeletonization via the realization of the fire front's propagation and extinction in digital binary shapes.IEEE Trans. Patt. Anal Mach. Intell., 11(10): 1076–1089.Google Scholar
  37. Yeshurun, Y. and Schwartz, E.L., 1989. Shape description with a space-variant sensor: Algorithm for scan-path, fusion, and convergence over multiple scans,IEEE Trans. Patt. Anal. Mach. Intell.,11(11):1217–1222.Google Scholar
  38. Yuille, A. and Leyton, M., 1990. 3d symmetry-curvature duality theorems,J. Comput. Vis. Graphics, Image Process. 52:124–140.Google Scholar
  39. Zabrodsky, H., Peleg, S., and Avnir, D., 1992. A measure of symmetry based on shape similarity,Proc. Conf. Comput. Vis. Patt. Recog., Champaign, IL, June.Google Scholar
  40. Zucker, S.W., Dobbins, A., and Iverson, L., 1989. Two stages of curve detection suggest two styles of visual computation,Neural Computation 1:68–81.Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • Daniel Reisfeld
    • 1
  • Haim Wolfson
    • 1
  • Yehezkel Yeshurun
    • 1
  1. 1.Computer Science Department, Sackler Faculty of Exact SciencesTel Aviv UniversityIsrael

Personalised recommendations