UNITY IN DIVERSITY: DISCOVERING TOPICS FROM WORDS - Information Theoretic Co-clustering for Visual Categorization

Ashish Gupta, Richard Bowden

Abstract

This paper presents a novel approach to learning a codebook for visual categorization, that resolves the key issue of intra-category appearance variation found in complex real world datasets. The codebook of visual-topics (semantically equivalent descriptors) is made by grouping visual-words (syntactically equivalent descriptors) that are scattered in feature space. We analyze the joint distribution of images and visual-words using information theoretic co-clustering to discover visual-topics. Our approach is compared with the standard ‘Bagof-Words’ approach. The statistically significant performance improvement in all the datasets utilized (Pascal VOC 2006; VOC 2007; VOC 2010; Scene-15) establishes the efficacy of our approach.

References

  1. Boiman, O., Shechtman, E., and Irani, M. (2008). In defense of nearest-neighbor based image classification. In CVPR08, pages 1-8.
  2. Table 4: Expt. Topic Codebook Size: Comparison of information theoretic co-clustering in terms of performance gain over the BoW model. The dataset utilized is Pacal VOC2010. The graphs show the F1-score for each technique across the visual categories in the dataset. The codebook sizes are 50, 100, 500, 1000, and 5000.
  3. Cheng, Y. and Church, G. M. (2000). Biclustering of expression data. In Proc Int Conf Intell Syst Mol Biol., volume 8, pages 93-103.
  4. Csurka, G., Dance, C. R., Fan, L., Willamowski, J., and Bray, C. (2004). Visual categorization with bags of keypoints. In In Workshop on Statistical Learning in Computer Vision, ECCV, pages 1-22.
  5. Dhillon, I. S. (2001). Co-clustering documents and words using bipartite spectral graph partitioning. In Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining, KDD 7801, pages 269-274, New York, NY, USA. ACM.
  6. Dhillon, I. S., Mallela, S., and Modha, D. S. (2003). Information-theoretic co-clustering. In In KDD, pages 89-98. ACM Press.
  7. Everingham, M. (2006). The PASCAL Visual Object Classes Challenge 2006 (VOC2006) Results. http:// www.pascal-network.org/ challenges/ VOC/ voc2006/ results.pdf.
  8. Everingham, M. (2007). The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results. http:// www.pascal-network.org/ challenges/ VOC/ voc2007/ workshop/index.html.
  9. Everingham, M. (2010). The PASCAL Visual Object Classes Challenge 2010 (VOC2010) Results. http:// www.pascal-network.org/ challenges/ VOC/ voc2010/ workshop/index.html.
  10. Fei-fei, L. (2005). A bayesian hierarchical model for learning natural scene categories. In In Proc. CVPR, pages 524-531.
  11. Gupta, A. and Bowden, R. (2011). Evaluating dimensionality reduction techniques for visual category recognition using renyi entropy. In In Proc. of European Signal Processing Conf., pages 913-917.
  12. Hartigan, J. A. (1972). Direct clustering of a data matrix. Journal of the American Statistical Association, 67(337):pp. 123-129.
  13. Horster, E., Greif, T., Lienhart, R., and Slaney, M. (2008). Comparing local feature descriptors in plsa-based image models. In Proc. of the 30th DAGM symposium, pages 446-455, Berlin, Heidelberg.
  14. Ke, Y. and Sukthankar, R. (2004). Pca-sift: a more distinctive representation for local image descriptors. In CVPR04, pages II: 506-513.
  15. Liu, J. and Shah, M. (2007). Scene modeling using coclustering. In In Proc. ICCV, pages 1-7.
  16. Lowe, D. (2004). Distinctive image features from scaleinvariant keypoints. International Journal on Computer Vision, 60(2):91-110.
  17. Rycroft, C. H., Grest, G. S., Landry, J. W., and Bazant, M. Z. (2006). Analysis of granular flow in a pebblebed nuclear reactor. Physical Review E, 74.
Download


Paper Citation


in Harvard Style

Gupta A. and Bowden R. (2012). UNITY IN DIVERSITY: DISCOVERING TOPICS FROM WORDS - Information Theoretic Co-clustering for Visual Categorization . In Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2012) ISBN 978-989-8565-03-7, pages 628-633. DOI: 10.5220/0003861206280633


in Bibtex Style

@conference{visapp12,
author={Ashish Gupta and Richard Bowden},
title={UNITY IN DIVERSITY: DISCOVERING TOPICS FROM WORDS - Information Theoretic Co-clustering for Visual Categorization},
booktitle={Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2012)},
year={2012},
pages={628-633},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003861206280633},
isbn={978-989-8565-03-7},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2012)
TI - UNITY IN DIVERSITY: DISCOVERING TOPICS FROM WORDS - Information Theoretic Co-clustering for Visual Categorization
SN - 978-989-8565-03-7
AU - Gupta A.
AU - Bowden R.
PY - 2012
SP - 628
EP - 633
DO - 10.5220/0003861206280633