FAST MULTI-CLASS IMAGE ANNOTATION WITH RANDOM SUBWINDOWS AND MULTIPLE OUTPUT RANDOMIZED TREES

Marie Dumont, Raphaël Marée, Louis Wehenkel, Pierre Geurts

2009

Abstract

This paper addresses image annotation, i.e. labelling pixels of an image with a class among a finite set of predefined classes. We propose a new method which extracts a sample of subwindows from a set of annotated images in order to train a subwindow annotation model by using the extremely randomized trees ensemble method appropriately extended to handle high-dimensional output spaces. The annotation of a pixel of an unseen image is done by aggregating the annotations of its subwindows containing this pixel. The proposed method is compared to a more basic approach predicting the class of a pixel from a single window centered on that pixel and to other state-of-the-art image annotation methods. In terms of accuracy, the proposed method significantly outperforms the basic method and shows good performances with respect to the state-of-the-art, while being more generic, conceptually simpler, and of higher computational efficiency than these latter.

References

  1. Amit, Y. and Geman, D. (1997). Shape quantization and recognition with randomized trees. Neural computation, 9:1545-1588.
  2. Bertelli, L., Byun, J., and Manjunath, B. S. (2007). A variational approach to exploit prior information in objectbackground segregation: application to retinal images. In ICIP.
  3. Borenstein, E. and Malik, J. (2006). Shape guided object segmentation. In CVPR, pages 969-976.
  4. Bosch, A., Zisserman, A., and Munoz, X. (2007). Image classification using random forests and ferns. In Proc. ICCV.
  5. Breiman, L., Friedman, J., Olshen, R., and Stone, C. (1984). Classification and Regression Trees. Wadsworth and Brooks, Monterey, CA.
  6. Cour, T. and Shi, J. (2007). Recognizing objects by piecing together the segmentation puzzle. In CVPR.
  7. Dollar, P., Tu, Z., and Belongie, S. (2006). Supervised learning of edges and object boundaries. In CVPR, pages 1964-1971.
  8. Dumont, M., Marée, R., Wehenkel, L., and Geurts, P. (2009). http://www.montefiore.ulg.ac.be/services/stochastic/ pubs/2009/DMWG09/.
  9. Ernst, D., Marée, R., and Wehenkel, L. (2006). Reinforcement learning with raw image pixels as input state. In Proc. International Workshop on Intelligent Computing in Pattern Analysis/Synthesis, volume 4153 of Lecture Notes in Computer Science, pages 446-454.
  10. Geurts, P., Ernst, D., and Wehenkel, L. (2006a). Extremely randomized trees. Machine Learning, 36(1):3-42.
  11. Geurts, P., Wehenkel, L., and d Alché-Buc, F. (2006b). Kernelizing the output of tree-based methods. In ICML, pages 345-352.
  12. He, X., Zemel, R., and Carreira-Perpinan, M. (2004). Multiscale conditional random fields for image labelling. In CVPR, volume 2, pages 695-702.
  13. He, X., Zemel, R., and Ray, D. (2006). Learning and incorporating top-down cues in image segmentation. In ECCV, pages 338-351.
  14. Jodogne, S., Briquet, C., and Piater, J. (2006). Approximate policy iteration for closed-loop learning of visual tasks. In Proc. of the 17th European Conference on Machine Learning, volume 4212, pages 222-226.
  15. Lepetit, V. and Fua, P. (2006). Keypoint recognition using randomized trees. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(9):1465-1479.
  16. Marée, R., Geurts, P., Piater, J., and Wehenkel, L. (2005). Random subwindows for robust image classification. In CVPR, volume 1, pages 34-40.
  17. Marée, R., Geurts, P., Visimberga, G., Piater, J., and Wehenkel, L. (2003). An empirical comparison of machine learning algorithms for generic image classification. In Proc. SGAI-AI, pages 169-182.
  18. Marée, R., Geurts, P., and Wehenkel, L. (2007). Contentbased image retrieval by indexing random subwindows with randomized trees. In Proc. ACCV, volume 4844 of LNCS, pages 611-620.
  19. Meurie, C., Lebrun, G., Lezoray, O., and Elmoataz, A. (2003). A comparison of supervised pixels-based color image segmentation methods. application in cancerology. WSEAS Transactions on computers, special issue on SSIP'03, 2:739-744.
  20. Moosmann, F., Triggs, B., and Jurie, F. (2006). Randomized clustering forests for building fast and discriminative visual vocabularies. In Neural Information Processing Systems (NIPS).
  21. Nowak, E. and Jurie, F. (2007). Learning visual similarity measures for comparing never seen objects. In Proc. IEEE CVPR.
  22. Nowak, E., Jurie, F., and Triggs, B. (2006). Sampling strategies for bag-of-features image classification. In Proc. ECCV, pages 490-503.
  23. Philbin, J., Chum, O., Isard, M., Sivic, J., and Zisserman, A. (2007). Object retrieval with large vocabularies and fast spatial matching. Proc. IEEE CVPR.
  24. Schroff, F., Criminisi, A., and Zisserman, A. (2008). Object class segmentation using random forests. In Proc. of the British Machine Vision Conference (BMVC).
  25. Shotton, J., Johnson, M., and Cipolla, R. (2008). Semantic texton forests for image categorization and segmentation. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
  26. Shotton, J., Winn, J., Rother, C., and Criminisi, A. (2006). Textonboost: Joint appearance, shape and context modeling for multi-class object recognition and segmentation. In ECCV, pages I: 1-15.
  27. Shotton, J., Winn, J., Rother, C., and Criminisi, A. (2007). Textonboost for image understanding: Multi-class object recognition and segmentation by jointly modeling appearance, shape and context. In International Journal on Computer Vision.
  28. Verbeek, J. and Triggs, B. (2007). Scene segmentation with conditional random fields learned from partially labeled images. In Proc. NIPS.
  29. Yin, P., Criminisi, A., Winn, J., and Essa, I. (2007). Treebased classifiers for bilayer video segmentation. In Proc. CVPR.
  30. Yu, S. X. and Shi, J. (2003). Object-specific figure-ground segregation. In CVPR, volume 2, pages 39-45.
Download


Paper Citation


in Harvard Style

Dumont M., Marée R., Wehenkel L. and Geurts P. (2009). FAST MULTI-CLASS IMAGE ANNOTATION WITH RANDOM SUBWINDOWS AND MULTIPLE OUTPUT RANDOMIZED TREES . In Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2009) ISBN 978-989-8111-69-2, pages 196-203. DOI: 10.5220/0001800001960203


in Bibtex Style

@conference{visapp09,
author={Marie Dumont and Raphaël Marée and Louis Wehenkel and Pierre Geurts},
title={FAST MULTI-CLASS IMAGE ANNOTATION WITH RANDOM SUBWINDOWS AND MULTIPLE OUTPUT RANDOMIZED TREES},
booktitle={Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2009)},
year={2009},
pages={196-203},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001800001960203},
isbn={978-989-8111-69-2},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Fourth International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2009)
TI - FAST MULTI-CLASS IMAGE ANNOTATION WITH RANDOM SUBWINDOWS AND MULTIPLE OUTPUT RANDOMIZED TREES
SN - 978-989-8111-69-2
AU - Dumont M.
AU - Marée R.
AU - Wehenkel L.
AU - Geurts P.
PY - 2009
SP - 196
EP - 203
DO - 10.5220/0001800001960203