Advertisement

Gated Boltzmann Machine in Texture Modeling

  • Tele Hao
  • Tapani Raiko
  • Alexander Ilin
  • Juha Karhunen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7553)

Abstract

In this paper, we consider the problem of modeling complex texture information using undirected probabilistic graphical models. Texture is a special type of data that one can better understand by considering its local structure. For that purpose, we propose a convolutional variant of the Gaussian gated Boltzmann machine (GGBM) [12], inspired by the co-occurrence matrix in traditional texture analysis. We also link the proposed model to a much simpler Gaussian restricted Boltzmann machine where convolutional features are computed as a preprocessing step. The usefulness of the model is illustrated in texture classification and reconstruction experiments.

Keywords

Gated Boltzmann Machine Texture Analysis Deep Learning Gaussian Restricted Boltzmann Machine 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Caputo, B., Hayman, E., Mallikarjuna, P.: Class-Specific Material Categorisation. In: Int. Conf. on Computer Vision, pp. 1597–1604 (2005)Google Scholar
  2. 2.
    Cho, K., Raiko, T., Ilin, A.: Gaussian-Bernoulli Deep Boltzmann Machine. In: NIPS 2011 Workshop on Deep Learning and Unsupervised Feature Learning (2011)Google Scholar
  3. 3.
    Cho, K., Ilin, A., Raiko, T.: Improved Learning of Gaussian-Bernoulli Restricted Boltzmann Machines. In: Int. Conf. on Artifical Neural Networks, pp. 10–17 (2011)Google Scholar
  4. 4.
    Cireşan, D.C., Meier, U., Gambardella, L.M., Schmidhuber, J.: Deep, Big, Simple Neural Nets for Handwritten Digit Recognition. Neural Comput. 22(12) (2010)Google Scholar
  5. 5.
    Fan, R.E., Chang, K.W., Hsieh, C.J., Wang, X.R., Lin, C.J.: LIBLINEAR: A Library for Large Linear Classification. JMLR, 1871–1874 (2008)Google Scholar
  6. 6.
    Haralick, R.M., Shanmugam, K., Dinstein, I.: Textural Features for Image Classification. IEEE Trans. Syst., Man, Cybern. 3(6), 610–621 (1973)CrossRefGoogle Scholar
  7. 7.
    Hinton, G., Salakhutdinov, R.R.: Reducing the dimensionality of data with neural networks. Science 313(5786), 504–507 (2006)CrossRefzbMATHMathSciNetGoogle Scholar
  8. 8.
    Hinton, G., Salakhutdinov, R.: Discovering Binary Codes for Documents by Learning Deep Generative Models. Topics in Cognitive Science 3(1), 74–91 (2010)CrossRefGoogle Scholar
  9. 9.
    Kivinen, J., Williams, C.: Multiple Texture Boltzmann Machines. JMLR W&CP 22, 638–646 (2012)Google Scholar
  10. 10.
    Lee, H., Grosse, R., Ranganath, R., Ng, A.Y.: Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations. In: Int. Conf. Machine Learning, p. 77 (2009)Google Scholar
  11. 11.
    Liu, L., Fieguth, P.: Texture Classification from Random Features. IEEE Trans. Pattern Anal. Mach. Intell. 34(3), 574–586 (2012)CrossRefGoogle Scholar
  12. 12.
    Memisevic, R., Hinton, G.E.: Learning to Represent Spatial Transformations with Factored Higher-Order Boltzmann Machines. Neural Comput. 22(6), 1473–1492 (2010)CrossRefzbMATHGoogle Scholar
  13. 13.
    Ranzato, M., Krizhevsky, A., Hinton, G.E.: Factored 3-Way Restricted Boltzmann Machines For Modeling Natural Images. JMLR W&CP 9, 621–628 (2010)Google Scholar
  14. 14.
    Varma, M., Zisserman, A.: A Statistical Approach to Material Classification Using Image Patch Exemplars. IEEE Trans. Pattern Anal. Mach. Intell. 31(11), 2032–2047 (2009)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Tele Hao
    • 1
  • Tapani Raiko
    • 1
  • Alexander Ilin
    • 1
  • Juha Karhunen
    • 1
  1. 1.Department of Information and Computer ScienceAalto UniversityEspooFinland

Personalised recommendations