Spatio-chromatic Image Content Descriptors and Their Analysis Using Extreme Value Theory

  • Vasileios Zografos
  • Reiner Lenz
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6688)


We use the theory of group representations to construct very fast image descriptors that split the vector space of local RGB distributions into small group-invariant subspaces. These descriptors are group theoretical generalizations of the Fourier Transform and can be computed with algorithms similar to the FFT. Because of their computational efficiency they are especially suitable for retrieval, recognition and classification in very large image datasets. We also show that the statistical properties of these descriptors are governed by the principles of the Extreme Value Theory (EVT). This enables us to work directly with parametric probability distribution models, which offer a much lower dimensionality and higher resolution and flexibility than histogram representations. We explore the connection to EVT and analyse the characteristics of these descriptors from a probabilistic viewpoint with the help of large image databases.


Weibull Distribution Dihedral Group Extreme Value Theory Large Image Database Image Search Engine 
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.
    Everingham, M., Van Gool, L., Williams, C.K.I., Winn, J., Zisserman, A.: The Pascal visual object classes (voc) challenge. IJCV 88, 303–338 (2010)CrossRefGoogle Scholar
  2. 2.
    Fässler, A., Stiefel, E.L.: Group theoretical methods and their applications. Birkhäuser, Boston (1992)CrossRefzbMATHGoogle Scholar
  3. 3.
    Lenz, R.: Investigation of receptive fields using representations of dihedral groups. Journal of Visual Communication and Image Representation 6, 209–227 (1995)CrossRefGoogle Scholar
  4. 4.
    Lenz, R., Bui, T.H., Takase, K.: A group theoretical toolbox for color image operators. In: ICIP, vol. 3, pp. 557–560 (2005)Google Scholar
  5. 5.
  6. 6.
    Gumbel, E.J.: Statistics of Extremes. Columbia University Press, New York (1958)zbMATHGoogle Scholar
  7. 7.
    Bertin, E., Clusel, M.: Generalised extreme value statistics and sum of correlated variables. Journal of Physics A: Mathematical and General 39 (2006)Google Scholar
  8. 8.
    Rinne, H.: The Weibull Distribution: A Handbook. CRC Press, Boca Raton (2008)CrossRefzbMATHGoogle Scholar
  9. 9.
    Kaplan, E.L., Meier, P.: Nonparametric estimation from incomplete observations. J. Amer. Statist. Assn. 53, 457–481 (1958)MathSciNetCrossRefzbMATHGoogle Scholar
  10. 10.
    Li, Y., Shapiro, L., Bilmes, J.: A generative/discriminative learning algorithm for image classification. In: ICCV, vol. 2, pp. 1605–1612 (2005)Google Scholar
  11. 11.
    Solli, M., Lenz, R.: Emotion related structures in large image databases. In: ACM CIVR, pp. 398–405 (2010)Google Scholar
  12. 12.
    Yanulevskaya, V., Geusebroek, J.M.: Significance of the Weibull distribution and its sub-models in natural image statistics. In: VISAPP, vol. 1, pp. 355–362 (2009)Google Scholar
  13. 13.
    Gijsenij, A., Gevers, T.: Color constancy using natural image statistics and scene semantics. IEEE PAMI 99 (2010)Google Scholar
  14. 14.
    Geusebroek, J.-M., Smeulders, A.W.M.: Fragmentation in the vision of scenes. In: ICCV, pp. 130–135 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Vasileios Zografos
    • 1
  • Reiner Lenz
    • 1
  1. 1.Computer Vision LaboratoryLinköping UniversitySweden

Personalised recommendations