Hierarchical Top Down Enhancements of Robust PCA

  • Georg Langs
  • Horst Bischof
  • Walter G. Kropatsch
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)


In this paper we deal with performance improvement of robust PCA algorithms by replacing regular subsampling of images by an irregular image pyramid adapted to the expected image content. The irregular pyramid is a structure built based on knowledge gained from the training set of images. It represents different regions of the image with different level of detail, depending on their importance for reconstruction. This strategy enables us to improve reconstruction results and therefore the recognition significantly. The training algorithm works on the data necessary to perform robust PCA and therefore requires no additional input.


Input Image Training Image Reconstruction Error Full Resolution Neighborhood Relation 
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.
    Lawrence W. Stark and Claudio M. Privitera. Top-down and bottom-up image processing. In Int. Conf. On Neural Networks, volume 4, pages 2294–2299, 1997.Google Scholar
  2. 2.
    D.A. Chernyak and L.W. Stark. Top-down guided eye movements. SMC-B, 31(4):514–522, August 2001.Google Scholar
  3. 3.
    Claudio M. Privitera and Lawrence W. Stark. Algorithms for defining visual regions-of-interest: Comparison with eye fixations. IEEE Trans. on PAMI, 22(9), 2000.Google Scholar
  4. 4.
    Aleš Leonardis and Horst Bischof. Robust recognition using eigenimages. CVIU, 78:99–118, 2000.Google Scholar
  5. 5.
    Walter G. Kropatsch. Irregular pyramids. Technical Report PRIP-TR-5, Institute for Automation, Pattern Recognition and Image Processing Group, University of Technology, Vienna.Google Scholar
  6. 6.
    Peter Meer. Stochastic image pyramids. CVGIP, 45:269–294, 1989.Google Scholar
  7. 7.
    Jean-Michel Jolion. Data driven decimation of graphs. In Proc. Of GbR’01, 3rd IAPR Int. Workshop on Graph Based Representations, pages 105–114, 2001.Google Scholar
  8. 8.
    Georg Langs, Horst Bischof, and Walter G. Kropatsch. Irregular image pyramids and robust appearance-based object recognition. Technical Report PRIP-TR-67, Institute for Automation, Pattern Recognition and Image Processing Group, University of Technology, Vienna.Google Scholar
  9. 9.
    S.A. Nene, S.K. Nayar, and H. Murase. Columbia object image library (COIL20). Technical Report CUCS-005-96, Columbia University, New York, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Georg Langs
    • 1
  • Horst Bischof
    • 2
  • Walter G. Kropatsch
    • 1
  1. 1.Pattern Recognition and Image Processing Group 183/2 Institute for Computer Aided AutomationVienna University of TechnologyViennaAustria
  2. 2.Institute for Computer Graphics and VisionGrazAustria

Personalised recommendations