Learning to recognize faces from examples

  • Shimon Edelman
  • Daniel Reisfeld
  • Yechezkel Yeshurun
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 588)


We describe an implemented system that learns to recognize human faces under varying pose and illumination conditions. The system relies on symmetry operations to detect the eyes and the mouth in a face image, uses the locations of these features to normalize the appearance of the face, performs simple but effective dimensionality reduction by a convolution with a set of Gaussian receptive fields, and subjects the vector of activities of the receptive fields to a Radial Basis Function interpolating classifier. The performance of the system compares favorably with the state of the art in machine recognition of faces.


Face Recognition Receptive Field Object Recognition Face Image False Alarm Rate 
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.
    R. Brunelli and T. Poggio. HyperBF networks for real object recognition. In Proceedings IJCAI, pages 1278–1284, Sydney, Australia, 1991.Google Scholar
  2. 2.
    R. O. Duda and P. E. Hart. Pattern classification and scene analysis. Wiley, New York, 1973.Google Scholar
  3. 3.
    S. Edelman and T. Poggio. Bringing the Grandmother back into the picture: a memorybased view of object recognition. A.I. Memo No. 1181, AI Lab, MIT, 1990. to appear in Int. J. Pattern Recog. Artif. Intell.Google Scholar
  4. 4.
    N. Intrator, J. I. Gold, H. H. Bülthoff, and S. Edelman. Three-dimensional object recognition using an unsupervised neural network: understanding the distinguishing features. In D. Touretzky, editor, Neural Information Processing Systems, volume 4. Morgan Kaufmann, San Mateo, CA, 1992. to appear.Google Scholar
  5. 5.
    M. Kirby and L. Sirovich. Application of the Karhunen-Loève procedure for characterization of human faces. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(1):103–108, 1990.Google Scholar
  6. 6.
    T. Poggio. A theory of how the brain might work. Cold Spring Harbor Symposia on Quantitative Biology, LV:899–910, 1990.Google Scholar
  7. 7.
    T. Poggio and S. Edelman. A network that learns to recognize three-dimensional objects. Nature, 343:263–266, 1990.PubMedGoogle Scholar
  8. 8.
    T. Poggio, M. Fahle, and S. Edelman. Synthesis of visual modules from examples: learning hyperacuity. A.I. Memo No. 1271, AI Lab, MIT, 1991. to appear in CVGIP B, 1992.Google Scholar
  9. 9.
    T. Poggio and F. Girosi. Regularization algorithms for learning that are equivalent to multilayer networks. Science, 247:978–982, 1990.Google Scholar
  10. 10.
    D. Reisfeld, H. Wolfson, and Y. Yeshurun. Detection of interest points using symmetry. In Proceedings of the 3rd International Conference on Computer Vision, pages 62–65, Tokyo, 1990. IEEE, Washington, DC.Google Scholar
  11. 11.
    D. Reisfeld and Y. Yeshurun. Robust Detection of Facial Features by Generalized Symmetry 1991. in preparation.Google Scholar
  12. 12.
    M. Turk and A. Pentland. Eigenfaces for recognition. J. of Cognitive Neuroscience, 3:71–86, 1991.Google Scholar
  13. 13.
    S. Ullman. Aligning pictorial descriptions: an approach to object recognition. Cognition, 32:193–254, 1989.PubMedGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Shimon Edelman
    • 1
  • Daniel Reisfeld
    • 2
  • Yechezkel Yeshurun
    • 2
  1. 1.Dept. of Applied Mathematics and Computer ScienceThe Weizmann Institute of ScienceRehovotIsrael
  2. 2.Dept. of Computer ScienceTel Aviv UniversityTel AvivIsrael

Personalised recommendations