A Heuristic Based on the Intrinsic Dimensionality for Reducing the Number of Cyclic DTW Comparisons in Shape Classification and Retrieval Using AESA

  • Vicente Palazón-González
  • Andrés Marzal
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7626)

Abstract

Cyclic Dynamic Time Warping (CDTW) is a good dissimilarity of shape descriptors of high dimensionality based on contours, but it is computationally expensive. For this reason, to perform recognition tasks, a method to reduce the number of comparisons and avoid an exhaustive search is convenient. The Approximate and Eliminate Search Algorithm (AESA) is a relevant indexing method because of its drastic reduction of comparisons, however, this algorithm requires a metric distance and that is not the case of CDTW. In this paper, we introduce a heuristic based on the intrinsic dimensionality that allows to use CDTW and AESA together in classification and retrieval tasks over these shape descriptors. Experimental results show that, for descriptors of high dimensionality, our proposal is optimal in practice and significantly outperforms an exhaustive search, which is the only alternative for them and CDTW in these tasks.

Keywords

Cyclic strings cyclic sequences cyclic dynamic time warping shape classification shape retrieval intrinsic dimensionality metric spaces AESA 

References

  1. 1.
    Sankoff, D., Kruskal, J. (eds.): Time Warps, String Edits, and Macromolecules: the Theory and Practice of Sequence Comparison. Addison-Wesley, Reading (1983)Google Scholar
  2. 2.
    Arica, N., Yarman-Vural, F.T.: BAS: a perceptual shape descriptor based on the beam angle statistics. Pattern Recognition Letters 24(9-10), 1627–1639 (2003)MATHCrossRefGoogle Scholar
  3. 3.
    Adamek, T., O’Connor, N.E.: A multiscale representation method for nonrigid shapes with a single closed contour. IEEE Trans. Circuits Syst. Video Techn. 14(5), 742–753 (2004)CrossRefGoogle Scholar
  4. 4.
    Alajlan, N., Rube, I.E., Kamel, M.S., Freeman, G.: Shape retrieval using triangle-area representation and dynamic space warping. Pattern Recognition 40(7), 1911–1920 (2007)MATHCrossRefGoogle Scholar
  5. 5.
    Belongie, S., Malik, J., Puzicha, J.: Shape matching and object recognition using shape contexts. IEEE Trans. Pattern Anal. Mach. Intell. 24(4), 509–522 (2002)CrossRefGoogle Scholar
  6. 6.
    Ling, H., Jacobs, D.W.: Shape classification using the inner-distance. IEEE Trans. Pattern Anal. Mach. Intell. 29(2), 286–299 (2007)CrossRefGoogle Scholar
  7. 7.
    Gopalan, R., Turaga, P., Chellappa, R.: Articulation-Invariant Representation of Non-planar Shapes - Supplementary Material. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part III. LNCS, vol. 6313, pp. 286–299. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  8. 8.
    Wang, J., Bai, X., You, X., Liu, W., Latecki, L.: Shape matching and classification using height functions. Pattern Recognition Letters (2011)Google Scholar
  9. 9.
    Keogh, E., Wei, L., Xi, X., Vlachos, M., Lee, S., Protopapas, P.: Supporting exact indexing of arbitrarily rotated shapes and periodic time series under Euclidean and warping distance measures. The VLDB Journal 18(3), 611–630 (2009)CrossRefGoogle Scholar
  10. 10.
    Marzal, A., Palazón, V., Peris, G.: Shape Retrieval Using Normalized Fourier Descriptors Based Signatures and Cyclic Dynamic Time Warping. In: Yeung, D.-Y., Kwok, J.T., Fred, A., Roli, F., de Ridder, D. (eds.) SSPR 2006 and SPR 2006. LNCS, vol. 4109, pp. 208–216. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  11. 11.
    Keogh, E.J., Ratanamahatana, C.A.: Exact indexing of dynamic time warping. Knowl. Inf. Syst. 7(3), 358–386 (2005)CrossRefGoogle Scholar
  12. 12.
    Vidal, E.: An algorithm for finding nearest neighbours in (approximately) constant average time. Pattern Recognition Letters 4(3), 145–157 (1986)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Vidal, E.: New formulation and improvements of the nearest-neighbour approximating and eliminating search algorithm (AESA). Pattern Recognition Letters 15(1), 1–7 (1994)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Vidal, E., Casacuberta, F., Rulot, H.M.: Is the DTW distance really a metric? An algorithm reducing the number of DTW comparisons in isolated word recognition. Speech Communication 4(4), 333–344 (1985)CrossRefGoogle Scholar
  15. 15.
    Casacuberta, F., Vidal, E., Rulot, H.: On the metric properties of dynamic time warping. IEEE Trans. Acoustics, Speech and Signal Processing ASSP-35(11), 1631 (1987)CrossRefGoogle Scholar
  16. 16.
    Vidal, E., Casacuberta, F., Benedi, J., Lloret, M.: On the verification of triangle inequality by dynamic time-warping dissimilarity measures. Speech Commun. 7(1), 67–79 (1988)CrossRefGoogle Scholar
  17. 17.
    Vidal, E., Rulot, H.M., Casacuberta, F., Benedi, J.M.: On the use of a metric-space search algorithm (AESA) for fast DTW-based recognition of isolated words. IEEE Trans. Acoustics, Speech and Signal Processing ASSP-36(5), 651 (1988)MATHCrossRefGoogle Scholar
  18. 18.
    Chávez, E., Navarro, G., Baeza-Yates, R., Marroquín, J.: Searching in metric spaces. ACM Computing Surveys (CSUR) 33(3), 273–321 (2001)CrossRefGoogle Scholar
  19. 19.
    Lemire, D.: Faster retrieval with a two-pass dynamic-time-warping lower bound. Pattern Recognition 42(9), 2169–2180 (2009)MATHCrossRefGoogle Scholar
  20. 20.
    Bober, M.: MPEG-7 visual shape descriptors. IEEE Trans. Circuits Syst. Video Techn. 11(6), 716–719 (2001)CrossRefGoogle Scholar
  21. 21.
    Sharvit, D., Chan, J., Tek, H., Kimia, B.B.: Symmetry-based indexing of image databases. In: Workshop on Content-Based Access of Image and Video Libraries, pp. 56–62 (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Vicente Palazón-González
    • 1
  • Andrés Marzal
    • 1
  1. 1.Dept. Llenguatges i Sistemes Informàtics and Institute of New Imaging TechnologiesUniversitat Jaume I de CastellóSpain

Personalised recommendations