Invariant Object Recognition with Slow Feature Analysis

  • Mathias Franzius
  • Niko Wilbert
  • Laurenz Wiskott
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5163)

Abstract

Primates are very good at recognizing objects independently of viewing angle or retinal position and outperform existing computer vision systems by far. But invariant object recognition is only one prerequisite for successful interaction with the environment. An animal also needs to assess an object’s position and relative rotational angle. We propose here a model that is able to extract object identity, position, and rotation angles, where each code is independent of all others. We demonstrate the model behavior on complex three-dimensional objects under translation and in-depth rotation on homogeneous backgrounds. A similar model has previously been shown to extract hippocampal spatial codes from quasi-natural videos. The rigorous mathematical analysis of this earlier application carries over to the scenario of invariant object recognition.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Rolls, E.T., Deco, G.: The Computational Neuroscience of Vision. Oxford University Press, New York (2002)Google Scholar
  2. 2.
    Franzius, M., Sprekeler, H., Wiskott, L.: Slowness and sparseness lead to place, head-diretion and spatial-view cells. Public Library of Science (PLoS) Computational Biology 3(8), 166 (2007)MathSciNetGoogle Scholar
  3. 3.
    Picard, R., Graczyk, C., Mann, S., Wachman, J., Picard, L., Campbell, L.: Vision texture (2002), http://vismod.media.mit.edu/vismod/imagery/VisionTexture/vistex.html
  4. 4.
    Toucan Corporation: Toucan virtual museum (2005), http://toucan.web.infoseek.co.jp/3DCG/3ds/FishModelsE.html
  5. 5.
    Wiskott, L., Sejnowski, T.: Slow feature analysis: Unsupervised learning of invariances. Neural Computation 14(4), 715–770 (2002)MATHCrossRefGoogle Scholar
  6. 6.
    Berkes, P., Wiskott, L.: Slow feature analysis yields a rich repertoire of complex cell properties. Journal of Vision 5(6), 579–602 (2005), http://journalofvision.org/5/6/9/ CrossRefGoogle Scholar
  7. 7.
    Hashimoto, W.: Quadratic forms in natural images. Network: Computation in Neural Systems 14(4), 765–788 (2003)CrossRefGoogle Scholar
  8. 8.
    Sprekeler, H., Michaelis, C., Wiskott, L.: Slowness: An objective for spike-timing-plasticity? PLoS Computational Biology 3(6), 112 (2007)CrossRefMathSciNetGoogle Scholar
  9. 9.
    Wiskott, L.: Slow feature analysis: A theoretical analysis of optimal free responses. Neural Computation 15(9), 2147–2177 (2003)MATHCrossRefGoogle Scholar
  10. 10.
    Berkes, P., Zito, T.: Modular toolkit for data processing (version 2.0) (2005), http://mdp-toolkit.sourceforge.net
  11. 11.
    Földiák, P.: Learning invariance from transformation sequences. Neural Computation 3, 194–200 (1991)CrossRefGoogle Scholar
  12. 12.
    Stone, J.V., Bray, A.: A learning rule for extracting spatio-temporal invariances. Network: Computation in Neural Systems 6, 429–436 (1995)MATHCrossRefGoogle Scholar
  13. 13.
    Kayser, C., Einhäuser, W., Dümmer, O., König, P., Körding, K.: Extracting slow subspaces from natural videos leads to complex cells. In: Dorffner, G., Bischof, H., Hornik, K. (eds.) ICANN 2001. LNCS, vol. 2130, pp. 1075–1080. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  14. 14.
    Rolls, E.T.: Neurophysiological mechanisms underlying face processing within and beyond the temporal cortical visual areas. Philosophical Transactions of the Royal Society 335, 11–21 (1992)CrossRefGoogle Scholar
  15. 15.
    Wallis, G., Rolls, E.T.: Invariant face and object recognition in the visual system. Progress in Neurobiology 51(2), 167–194 (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Mathias Franzius
    • 1
  • Niko Wilbert
    • 1
  • Laurenz Wiskott
    • 1
  1. 1.Institute for Theoretical BiologyHumboldt-Universität zu BerlinGermany

Personalised recommendations