Randomized Locality Sensitive Vocabularies for Bag-of-Features Model

  • Yadong Mu
  • Ju Sun
  • Tony X. Han
  • Loong-Fah Cheong
  • Shuicheng Yan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6313)


Visual vocabulary construction is an integral part of the popular Bag-of-Features (BOF) model. When visual data scale up (in terms of the dimensionality of features or/and the number of samples), most existing algorithms (e.g. k-means) become unfavorable due to the prohibitive time and space requirements. In this paper we propose the random locality sensitive vocabulary (RLSV) scheme towards efficient visual vocabulary construction in such scenarios. Integrating ideas from the Locality Sensitive Hashing (LSH) and the Random Forest (RF), RLSV generates and aggregates multiple visual vocabularies based on random projections, without taking clustering or training efforts. This simple scheme demonstrates superior time and space efficiency over prior methods, in both theory and practice, while often achieving comparable or even better performances. Besides, extensions to supervised and kernelized vocabulary constructions are also discussed and experimented with.


Random Forest Hash Function Visual Word Reproduce Kernel Hilbert Space Mean Average Precision 
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.
    Lowe, D.: Distinctive image features from scale-invariant keypoints. International journal of computer vision 60, 91–110 (2004)CrossRefGoogle Scholar
  2. 2.
    Beyer, K.S., Goldstein, J., Ramakrishnan, R., Shaft, U.: When is nearest neigh- bor meaningful? In: Beeri, C., Bruneman, P. (eds.) ICDT 1999. LNCS, vol. 1540, pp. 217–235. Springer, Heidelberg (1998)Google Scholar
  3. 3.
    Dasgupta, S.: Experiments with random projection. In: UAI, pp. 143–151 (2000)Google Scholar
  4. 4.
    Jurie, F., Triggs, B.: Creating efficient codebooks for visual recognition. ICCV, 604–610 (2005)Google Scholar
  5. 5.
    Nistér, D., Stewénius, H.: Scalable recognition with a vocabulary tree. In: CVPR, vol. (2), pp. 2161–2168 (2006)Google Scholar
  6. 6.
    Breiman, L.: Random forests. Machine Learning 45, 5–32 (2001)zbMATHCrossRefGoogle Scholar
  7. 7.
    Moosmann, F., Nowak, E., Jurie, F.: Randomized clustering forests for image classiffcation. IEEE Trans. Pattern Anal. 30, 1632–1646 (2008)CrossRefGoogle Scholar
  8. 8.
    Yang, L., Jin, R., Sukthankar, R., Jurie, F.: Unifying discriminative visual code-book generation with classifier training for object category recognition. In: CVPR (2008)Google Scholar
  9. 9.
    Lee, H., Battle, A., Raina, R., Ng, A.: Efficient sparse coding algorithms. In: NIPS, vol. 19, p. 801 (2007)Google Scholar
  10. 10.
    Wright, J., Yang, A., Ganesh, A., Sastry, S., Ma, Y.: Robust face recognition via sparse representation. PAMI, 210–227 (2009)Google Scholar
  11. 11.
    Cormen, T., Leiserson, C., Rivest, R., Stein, C.: Introduction to algorithms. The MIT press, Cambridge (2001)zbMATHGoogle Scholar
  12. 12.
    Charikar, M.: Similarity estimation techniques from rounding algorithms. In: STOC, pp. 380–388 (2002)Google Scholar
  13. 13.
    Andoni, A., Indyk, P.: Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions. ACM Commun. 51, 117–122 (2008)CrossRefGoogle Scholar
  14. 14.
    Kulis, B., Grauman, K.: Kernelized locality-sensitive hashing for scalable image search. In: ICCV (2009)Google Scholar
  15. 15.
    Bentley, J.L.: Multidimensional binary search trees used for associative searching. ACM Commun. 517, 509–517 (1975)CrossRefGoogle Scholar
  16. 16.
    Datar, M., Immorlica, N., Indyk, P., Mirrokni, V.: Locality-sensitive hashing scheme based on p-stable distributions. In: SCG, pp. 253–262 (2004)Google Scholar
  17. 17.
    Indyk, P., Motwani, R.: Approximate nearest neighbors: Towards removing the curse of dimensionality. In: STOC, pp. 604–613 (1998)Google Scholar
  18. 18.
    Andoni, A., Indyk, P.: Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions. ACM Commun. 51, 117–122 (2008)CrossRefGoogle Scholar
  19. 19.
    Scholkopf, B., Smola, A.J.: Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT Press, Cambridge (2001)Google Scholar
  20. 20.
    Geurts, P., Ernst, D., Wehenkel, L.: Extremely randomized trees. Machine Learning 63, 3–42 (2006)zbMATHCrossRefGoogle Scholar
  21. 21.
    Schuldt, C., Laptev, I., Caputo, B.: Recognizing human actions: a local SVM approach. In: ICPR (2004)Google Scholar
  22. 22.
    Laptev, I.: Marsza lek, M., Schmid, C., Rozenfeld, B.: Learning realistic human actions from movies (2008)Google Scholar
  23. 23.
    Nowak, E., Jurie, F., Triggs, B.: Sampling strategies for bag-of-features image classification. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3954, pp. 490–503. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  24. 24.
    Marée, R., Geurts, P., Piater, J.H., Wehenkel, L.: Random subwindows for robust image classification. In: CVPR, vol. (1), pp. 34–40 (2005)Google Scholar
  25. 25.
    Wu, X., Hauptmann, A.G., Ngo, C.W.: Practical elimination of near-duplicates from web video search. ACM Multimedia, 218–227 (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Yadong Mu
    • 1
  • Ju Sun
    • 1
    • 2
  • Tony X. Han
    • 3
  • Loong-Fah Cheong
    • 1
    • 2
  • Shuicheng Yan
    • 1
  1. 1.Electrical and Computer EngineeringNational University of SingaporeSingapore
  2. 2.Interactive & Digital Media InstituteNational University of SingaporeSingapore
  3. 3.Electrical and Computer EngineeringUniversity of MissouriColumbiaUSA

Personalised recommendations