Multimedia Tools and Applications

, Volume 69, Issue 2, pp 471–490 | Cite as

Boosted kernel for image categorization

  • Alexis Lechervy
  • Philippe-Henri Gosselin
  • Frédéric Precioso


Recent machine learning techniques have demonstrated their capability for identifying image categories using image features. Among these techniques, Support Vector Machines (SVM) present good results for example in Pascal Voc challenge 2011 [8], particularly when they are associated with a kernel function [28, 35]. However, nowadays image categorization task is very challenging owing to the sizes of benchmark datasets and the number of categories to be classified. In such a context, lot of effort has to be put in the design of the kernel functions and underlying semantic features. In the following of the paper we call semantic features the features describing the (semantic) content of an image. In this paper, we propose a framework to learn an effective kernel function using the Boosting paradigm to linearly combine weak kernels. We then use a SVM with this kernel to categorize image databases. More specifically, this method create embedding functions to map images in a Hilbert space where they are better classified. Furthermore, our algorithm benefits from boosting process to learn this kernel with a complexity linear with the size of the training set. Experiments are carried out on popular benchmarks and databases to show the properties and behavior of the proposed method. On the PASCAL VOC2006 database, we compare our method to simple early fusion, and on the Oxford Flowers databases we show that our method outperforms the best Multiple Kernel Learning (MKL) techniques of the literature.


Image categorization Kernel machines Boosting 


  1. 1.
    Awais M, Yan F, Mikolajczyk K, Kittler J (2011) Augmented kernel matrix vs classifier fusion for object recognition. In: British machine vision conference. BMVA Press, pp 60.1–60.11Google Scholar
  2. 2.
    Bach, FR, Lanckriet GRG (2004) Multiple kernel learning, conic duality, and the smo algorithm. In: International conference on machine learningGoogle Scholar
  3. 3.
    Cortes C (2009) Invited talk: can learning kernels help performance? In: International conference on machine learning, p 161Google Scholar
  4. 4.
    Cortes C, Mohri M, Rostamizadeh A (2010) Two-stage learning kernel algorithms. In: International conference on machine learningGoogle Scholar
  5. 5.
    Crammer K, Keshet J, Singer Y (2001) Kernel design using boosting. In: Advances in neural information processing systems. MIT Press, pp 537–544Google Scholar
  6. 6.
    Cristianini N, Shawe-Taylor J, Elisseff A, Kandola J (2001) On kernel target alignement. In: Advances in neural information processing systems. Vancouver, Canada, pp 367–373Google Scholar
  7. 7.
    Davis JV, Kulis B, Jain P, Sra S, Dhillon IS (2007) Information-theoretic metric learning. In: International conference on machine learning, vol 227. Corvalis, OregonGoogle Scholar
  8. 8.
    Everingham M, Van Gool L, Williams CKI, Winn J, Zisserman A (2011) The Pascal visual object classes challenge 2011 (VOC2011) results.
  9. 9.
    Figueiras-Vidal AR, Rokach L (2012) An exploration of research directions in machine ensemble theory and applications. In: European symposium on artificial neural networks, computational intelligence and machine learningGoogle Scholar
  10. 10.
    Friedman J, Hastie T, Tibshirani R (2000) Additive logistic regression: a statistical view of boosting. Annals Stat 2:337–374CrossRefMathSciNetGoogle Scholar
  11. 11.
    Gehler P, NowoziS S (2009) On feature combination for multiclass object classification. In: IEEE international conference on computer vision, pp 221–228Google Scholar
  12. 12.
    Gong Y, Lazebnik S, Gordo A, Perronnin F (2012) Iterative quantization: a procrustean approach to learning binary codes for large-scale image retrieval. IEEE Trans pattern Anal Mach Intell 99:1 (PrePrints)Google Scholar
  13. 13.
    Gosselin PH, Cord M (2006) Feature based approach to semi-supervised similarity learning. Pattern Recognit 39:1839–1851. Special issue on similarity-based pattern recognitionCrossRefMATHGoogle Scholar
  14. 14.
    Gosselin PH, Cord M, Philipp-Foliguet S (2008) Combining visual dictionary, kernel-based similarity and learning strategy for image category retrieval. Comput Vis Image Underst 110(3):403–417. Special issue on similarity matching in computer vision and multimediaCrossRefGoogle Scholar
  15. 15.
    Gosselin PH, Precioso F, Philipp-Foliguet S (2010) Incremental kernel learning for active image retrieval without global dictionaries. Pattern Recognit 44(10–11):2244–2254Google Scholar
  16. 16.
    Hazen TJ (2010) Multi-class svm optimization using mce training with application to topic identification. In: IEEE International conference on acoustics, speech, and signal processing, pp 5350–5353Google Scholar
  17. 17.
    Kawanabe M, Nakajima S, Binder A (2009) A procedure of adaptive kernel combination with kernel-target alignment for object classification. In: ACM international conference on image and video retrievalGoogle Scholar
  18. 18.
    Kloft M, Brefeld U, Sonnenburg S, Zien A (2011) Lp-norm multiple kernel learning. J Mach Learn Res 12:953–997MATHMathSciNetGoogle Scholar
  19. 19.
    Lanckriet GRG, Cristianini N, Bartlett N, El Ghaoui L, Jordan MI (2004) Learning the kernel matrix with semi-definite programming. Int J Mach Learn Res 5:27–72MATHGoogle Scholar
  20. 20.
    Madzarov G, Gjorgjevikj D, Chorbev I (2008) A multi-class svm classifier utilizing binary decision tree. Informatica 33(1):233–242MathSciNetGoogle Scholar
  21. 21.
    Nilsback ME, Zisserman A (2008) Automated flower classification over a large number of classes. In: Proceedings of the Indian conference on computer vision, graphics and image processingGoogle Scholar
  22. 22.
    Orabona F, Jie L (2011) Ultra-fast optimization algorithm for sparse multi kernel learning. In: Getoor L, Scheffer T (eds) Proceedings of the 28th international conference on machine learning (ICML-11), ICML ’11. New York, NY, USA, ACM, pp. 249–256Google Scholar
  23. 23.
    Picard D, Thome N, Cord M, Rakotomamonjy A (2012) Learning geometric combinations of gaussian kernels with alternating quasi-newton algorithm. In: European symposium on artificial neural networks, computational intelligence and machine learningGoogle Scholar
  24. 24.
    Rakotomamonjy A, Bach F, Canu S, Grandvalet Y (2008) Simplemkl. J Mach Learn Res 9:2491–2521MATHMathSciNetGoogle Scholar
  25. 25.
    Sánchez J, Perronnin F (2011) High-dimensional signature compression for large-scale image classification. In: CVPR, pp 1665–1672Google Scholar
  26. 26.
    Schölkopf B, Smola A (2002) Learning with Kernels. MIT Press, Cambridge, MAGoogle Scholar
  27. 27.
    Shawe-Taylor J, Cristianini N (2004) Kernel methods for Pattern Analysis. Cambridge University Press, ISBN 0-521-81397-2Google Scholar
  28. 28.
    Song Z, Chen Q, Huang ZY, Hua Y, Yan S (2011) Contextualizing object detection and classification. In: CVPR, pp 1585–1592Google Scholar
  29. 29.
    Sonnenburg S, Rtsh G, Schfer C, Schlkopf B (2006) Large scale multiple kernel learning. In J Mach Learn Res 7:1531–1565MATHGoogle Scholar
  30. 30.
    Tieu K, Viola P (2000) Boosting image retrieval. In: IEEE international conference on computer vision and pattern recognition, pp 228–235Google Scholar
  31. 31.
    Vapnik V (1982) Estimation of dependences based on empirical data. Springer-VerlagGoogle Scholar
  32. 32.
    Varma M, Babu BR (2009) More generality in efficient multiple kernel learning. In: International conference on machine learningGoogle Scholar
  33. 33.
    Vert R (2002) Designing a m-svm kernel for protein secondary structure prediction. Master’s thesis, DEA informatique de LorraineGoogle Scholar
  34. 34.
    Yu S, Tranchevent LC, De Moor B, Moreau Y (2011) Kernel-based data fusion for machine learning - Methods and applications in bioinformatics and text mining, vol 345. Studies in computational intelligence. SpringerGoogle Scholar
  35. 35.
    Yu Y, Zhang J, Huang Y, Zheng S, Ren W, Wang C, Huang K, Tan T (2010) Object detection by context and boosted hog-lbp. In: ECCV workshop on PASCAL VOCGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Alexis Lechervy
    • 1
  • Philippe-Henri Gosselin
    • 1
  • Frédéric Precioso
    • 2
  1. 1.ETIS/ENSEA - Université de Cergy-Pontoise - CNRSCergy-PontoiseFrance
  2. 2.I3S, UMR7271, UNS CNRSSophia AntipolisFrance

Personalised recommendations