Sequential Gaussian Mixture Models for Two-Level Conditional Random Fields

  • Sergey Kosov
  • Franz Rottensteiner
  • Christian Heipke
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8142)

Abstract

Conditional Random Fields are among the most popular techniques for image labelling because of their flexibility in modelling dependencies between the labels and the image features. This paper addresses the problem of efficient classification of partially occluded objects. For this purpose we propose a novel Gaussian Mixture Model based on a sequential training procedure, in combination with multi-level CRF-framework. Our approach is evaluated on urban aerial images. It is shown to increase the classification accuracy in occluded areas by up to 14,4%.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bishop, C.M.: Pattern Recognition and Machine Learning, 1st edn. Springer, New York (2006)Google Scholar
  2. 2.
    Boykov, Y., Jolly, M.: Interactive graph cuts for optimal boundary and region segmentation of objects in n-d images. In: Proc. ICCV., vol. I, pp. 105–112 (2001)Google Scholar
  3. 3.
    Bradski, G.: The OpenCV Library. Dr. Dobb’s Journal of Software Tools (2000)Google Scholar
  4. 4.
    Cramer, M.: The DGPF test on digital aerial camera evaluation - overview and test design. Photogrammetrie-Fernerkundung-Geoinformation 2(2010), 73–82 (2010)CrossRefGoogle Scholar
  5. 5.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proc. CVPR, pp. 886–893 (2005)Google Scholar
  6. 6.
    DGM: Direct graphical models library (2013), http://research.project-10.de/dgm/
  7. 7.
    Fearnhead, P.: Particle filters for mixture models with an unknown number of components. Statistics and Computing 14(1), 11–21 (2004)MathSciNetCrossRefGoogle Scholar
  8. 8.
    Hinz, S., Baumgartner, A.: Automatic extraction of urban road networks from multi-view aerial imagery. ISPRS J. Photogramm. & Rem. Sens. 58, 83–98 (2003)CrossRefGoogle Scholar
  9. 9.
    Hirschmüller, H.: Stereo processing by semiglobal matching and mutual information. IEEE Trans. Pattern Anal. Mach. Intell. 30(2), 328–341 (2008)CrossRefGoogle Scholar
  10. 10.
    Kosov, S., Kohli, P., Rottensteiner, F., Heipke, C.: A two-layer conditional random field for the classification of partially occluded objects arXiv:1307.3043 [cs.CV] (2013), http://arxiv.org/abs/1307.3043
  11. 11.
    Kumar, S., Hebert, M.: A hierarchical field framework for unified context-based classification. In: Proc. ICCV, pp. 1284–1291 (2005)Google Scholar
  12. 12.
    Kumar, S., Hebert, M.: Discriminative Random Fields. Int. J. Comput. Vis. 68(2), 179–201 (2006), http://www.springerlink.com/index/10.1007/s11263-006-7007-9 CrossRefGoogle Scholar
  13. 13.
    Leibe, B., Leonardis, A., Schiele, B.: Robust object detection with interleaved categorization and segmentation. Int. J. Comput. Vis. 77, 259–289 (2008)CrossRefGoogle Scholar
  14. 14.
    McLachlan, G., Krishnan, T.: The EM algorithm and extensions, 2nd edn. Wiley series in probability and statistics. Wiley, Hoboken (2008)CrossRefMATHGoogle Scholar
  15. 15.
    Ravanbakhsh, M., Heipke, C., Pakzad, K.: Road junction extraction from high resolution aerial imagery. Photogrammetric Record 23, 405–423 (2008)CrossRefGoogle Scholar
  16. 16.
    Reynolds, D.A.: Gaussian mixture models. In: Encyclopedia of Biometrics, pp. 659–663. Springer, US (2009)Google Scholar
  17. 17.
    Rutzinger, M., Rottensteiner, F., Pfeifer, N.: A comparison of evaluation techniques for building extraction from airborne laser scanning. JSTARS 2(1), 11–20 (2009)Google Scholar
  18. 18.
    Schindler, K.: An overview and comparison of smooth labeling methods for land-cover classification. IEEE-TGARS 50, 4534–4545 (2012)Google Scholar
  19. 19.
    Schnitzspan, P., Fritz, M., Roth, S., Schiele, B.: Discriminative structure learning of hierarchical representations for object detection. In: CVPR, pp. 2238–2245 (2009)Google Scholar
  20. 20.
    Shotton, J., Winn, J., Rother, C., Criminisi, A.: Textonboost for image understanding: Multi-class object recognition and segmentation by jointly modeling texture, layout, and context. Int. J. Comput. Vis. 81, 2–23 (2009)CrossRefGoogle Scholar
  21. 21.
    Vishwanathan, S.V.N., Schraudolph, N.N., Schmidt, M.W., Murphy, K.P.: Accelerated training of conditional random fields with stochastic gradient methods. In: Proc. 23rd ICML, pp. 969–976 (2006)Google Scholar
  22. 22.
    Winn, J., Shotton, J.: The layout consistent random field for recognizing and segmenting partially occluded objects. In: Proc. CVPR (2006)Google Scholar
  23. 23.
    Yang, Y., Hallman, S., Ramanan, D., Fowlkes, C.: Layered object detection for multi-class segmentation. In: CVPR (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Sergey Kosov
    • 1
  • Franz Rottensteiner
    • 1
  • Christian Heipke
    • 1
  1. 1.Institute of Photogrammetry and GeoInformation (IPI)Leibniz Universität HanoverGermany

Personalised recommendations