Face Recognition in Real-Time Applications: A Comparison of Directed Enumeration Method and K-d Trees

  • Andrey V. Savchenko
Part of the Lecture Notes in Business Information Processing book series (LNBIP, volume 128)


The problem of face recognition with large database in real-time applications is discovered. The enhancement of HoG (Histogram of Gradients) algorithm with features mutual alignment is proposed to achieve better accuracy. The novel modification of directed enumeration method (DEM) using the ideas of the Best Bin First (BBF) search algorithm is introduced as an alternative to the nearest neighbor rule to prevent the brute force. We present the results of an experimental study in a problem of face recognition with FERET and Essex datasets. We compare the performance of our DEM modification with conventional BBF k-d trees in their well-known efficient implementation from OpenCV library. It is shown that the proposed method is characterized by increased computing efficiency (2-12 times in comparison with BBF) even in the most difficult cases where many neighbors are located at very similar distances. It is demonstrated that BBF cannot be used with our recognition algorithm as the latter is based on non-symmetric measure of similarity. However, we experimentally prove that our recognition algorithm improves recognition accuracy in comparison with classical HoG implementation. Finally, we show that this algorithm could be implemented efficiently if it is combined with the DEM.


Real-time object recognition HoG (histogram of gradients) directed enumeration method k-d tree Best Bin First 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Yi, M.: Abnormal Event Detection Method for ATM Video and Its Application. In: Lin, S., Huang, X. (eds.) CESM 2011, Part II. CCIS, vol. 176, pp. 186–192. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  2. 2.
    Nusimow, A.: Intelligent Video for Homeland Security Applications. In: Proceedings of the 7th Technologies for Homeland Security, pp. 139–144 (2007)Google Scholar
  3. 3.
    Zhao, W., Chellappa, R. (eds.): Face Processing: Advanced Modeling and Methods. Elsevier/Academic Press (2005)Google Scholar
  4. 4.
    Shan, C.: Face Recognition and Retrieval in Video. In: Schonfeld, D., Shan, C., Tao, D., Wang, L. (eds.) Video Search and Mining. SCI, vol. 287, pp. 235–260. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  5. 5.
    Beis, J., Lowe, D.G.: Shape indexing using approximate nearest-neighbour search in high dimensional spaces. In: Conference on Computer Vision and Pattern Recognition, Puerto Rico, pp. 1000–1006 (1997)Google Scholar
  6. 6.
    Savchenko, A.V.: Directed enumeration method in image recognition. Pattern Recognition 45(8), 2952–2961 (2012)CrossRefGoogle Scholar
  7. 7.
    Penteado, B.E., Marana, A.N.: A Video-Based Biometric Authentication for e-Learning Web Applications. In: Filipe, J., Cordeiro, J. (eds.) ICEIS 2009. LNBIP, vol. 24, pp. 770–779. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  8. 8.
    Arya, S., Mount, D.M., Netanyahu, N.S., Silverman, R., Wu, A.Y.: An optimal algorithm for approximate nearest neighbor searching. Journal of the ACM 45, 891–923 (1998)CrossRefGoogle Scholar
  9. 9.
    Liu, T., Moore, A.W., Gray, A.G., Yang, K.: An Investigation of Practical Approximate Nearest Neighbor Algorithms. In: NIPS (2004)Google Scholar
  10. 10.
    Bentley, J.L.: Multidimensional binary search trees used for associative searching. Communications of the ACM 18(9), 509–517 (1975)CrossRefGoogle Scholar
  11. 11.
    Shapiro, L., Stockman, G.: Computer vision, 752 p. Prentice Hall, Upper Saddle River (2001)Google Scholar
  12. 12.
    Theodoridis, S., Koutroumbas, C.: Pattern Recognition, 4th edn. Elsevier, Amsterdam (2009)Google Scholar
  13. 13.
    Kullback, S.: Information Theory and Statistics. Dover Pub., New York (1978)Google Scholar
  14. 14.
    Srisuk, S., Kurutach, W.: Face Recognition using a New Texture Representation of Face Images. In: Proceedings of Electrical Engineering Conference, pp. 1097–1102 (2003)Google Scholar
  15. 15.
    Dalal, N., Triggs, B.: Histograms of Oriented Gradients for Human Detection. In: International Conference on Computer Vision & Pattern Recognition, pp. 886–893 (2005)Google Scholar
  16. 16.
  17. 17.
  18. 18.
  19. 19.
    Roberts, L.: Machine Perception of 3-D Solids, Optical and Electro-optical Information Processing. MIT Press (1965)Google Scholar
  20. 20.
    Lowe, D.: Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 60(2), 91–110 (2004)CrossRefGoogle Scholar
  21. 21.
    Chow, C.K.: On optimum error and reject trade-off. IEEE Transactions on Information Theory 16, 41–46 (1970)CrossRefGoogle Scholar
  22. 22.
    Savchenko, A.V.: Image Recognition with a Large Database Using Method of Directed Enumeration Alternatives Modification. In: Kuznetsov, S.O., Ślęzak, D., Hepting, D.H., Mirkin, B.G. (eds.) RSFDGrC 2011. LNCS, vol. 6743, pp. 338–341. Springer, Heidelberg (2011)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Andrey V. Savchenko
    • 1
  1. 1.Higher School of EconomicsNational Research UniversityNizhniy NovgorodRussian Federation

Personalised recommendations