Random Subwindows for Robust Image Classification

Raphaël Marée, Pierre Geurts, Justus Piater, Louis Wehenkel
Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR 2005), Volume 1, page 34--40 - June 2005
Download the publication : maree-cvpr2005-slides.pdf [2.4Mo]   maree-cvpr2005.pdf [215Ko]  
We present a novel, generic image classification method based on a recent machine learning algorithm (ensembles of extremely randomized decision trees). Images are classified using randomly extracted subwindows that are suitably normalized to yield robustness to certain image transformations. Our method is evaluated on four very different, publicly available datasets (COIL-100, ZuBuD, ETH-80, WANG). Our results show that our automatic approach is generic and robust to illumination, scale, and viewpoint changes. An extension of the method is proposed to improve its robustness with respect to rotation changes.

See also

The java software PiXiT implements the method proposed in this paper.

BibTex references

@InProceedings\{MGPW05,
  author       = "Mar\'ee, Rapha{\"e}l and Geurts, Pierre and Piater, Justus and Wehenkel, Louis",
  title        = "Random Subwindows for Robust Image Classification",
  booktitle    = "Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR 2005)",
  volume       = "1",
  pages        = "34--40",
  month        = "June",
  year         = "2005",
  editor       = "Cordelia Schmid and Stefano Soatto and Carlo Tomasi",
  organization = "IEEE",
  keywords     = "machine learning",
  url          = "http://www.montefiore.ulg.ac.be/services/stochastic/pubs/2005/MGPW05"
}

Other publications in the database

» Raphaël Marée
» Pierre Geurts
» Justus Piater
» Louis Wehenkel