Advertisement

Shift-Invariant Dynamic Texture Recognition

  • Franco Woolfe
  • Andrew Fitzgibbon
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3952)

Abstract

We address the problem of recognition of natural motions such as water, smoke and wind-blown vegetation. Such dynamic scenes exhibit characteristic stochastic motions, and we ask whether the scene contents can be recognized using motion information alone. Previous work on this problem has considered only the case where the texture samples have sufficient overlap to allow registration, so that the visual content of the scene is very similar between examples. In this paper we investigate the recognition of entirely non-overlapping views of the same underlying motion, specifically excluding appearance-based cues.

We describe the scenes with time-series models—specifically multivariate autoregressive (AR) models—so the recognition problem becomes one of measuring distances between AR models. We show that existing techniques, when applied to non-overlapping sequences, have significantly lower performance than on static-camera data. We propose several new schemes, and show that some outperform the existing methods.

Keywords

Feature Vector Multivariate Time Series Dynamic Texture Texture Scene Univariate Time Series 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Johansson, G.: Visual perception of biological motion and a model for its analysis. Perception and Psychophysics 14, 201–211 (1973)CrossRefGoogle Scholar
  2. 2.
    Chetverikov, D., Peteri, R.: A brief survey of dynamic texture description and recognition. In: Intl. Conf. on Computer Recognition Systems, pp. 223–230 (2005)Google Scholar
  3. 3.
    Doretto, G., Chiuso, A., Wu, Y.N., Soatto, S.: Dynamic textures. IJCV 51, 91–109 (2003)CrossRefMATHGoogle Scholar
  4. 4.
    Soatto, S., Doretto, G., Wu, Y.N.: Dynamic textures. In: Proc. ICCV, pp. 439–446 (2001)Google Scholar
  5. 5.
    Fitzgibbon, A.W.: Stochastic rigidity: Image registration for nowhere-static scenes. In: Proc. ICCV, vol. 1, pp. 662–670 (2001)Google Scholar
  6. 6.
    Saisan, P., Doretto, G., Wu, Y.N., Soatto, S.: Dynamic texture recognition. In: Proc. CVPR, vol. 2, pp. 58–63 (2001)Google Scholar
  7. 7.
    Chan, A.B., Vasconcelos, N.: Probabilistic kernels for the classification of auto-regressive visual processes. In: Proc. CVPR, pp. 846–851 (2005)Google Scholar
  8. 8.
    Vidal, R., Ravichandran, A.: Optical flow estimation and segmentation of multiple moving dynamic textures. In: Proc. CVPR (2005)Google Scholar
  9. 9.
    Doretto, G., Soatto, S.: Towards plenoptic dynamic textures. In: Proc. Intl. Workshop on Texture Analysis and Synthesis (2003)Google Scholar
  10. 10.
    Polana, R., Nelson, R.: Detecting activities. In: Proc. CVPR, pp. 2–7 (1993)Google Scholar
  11. 11.
    Black, M.J.: Explaining optical flow events with parameterized spatio-temporal models. In: Proc. CVPR, pp. 1326–1332 (1999)Google Scholar
  12. 12.
    Niyogi, A.A.: Analyzing and recognizing walking figures in xyt. In: Proc. CVPR, pp. 469–474 (1994)Google Scholar
  13. 13.
    Laptev, I., Lindeberg, T.: Space-time interest points. In: Proc. ICCV, pp. 432–439 (2003)Google Scholar
  14. 14.
    Shechtman, E., Irani, M.: Space-time behavior based correlation. In: Proc. CVPR, pp. 405–412 (2005)Google Scholar
  15. 15.
    Yacoob, Y., Black, M.J.: Parameterized modeling and recognition of activities. In: CVIU, pp. 232–247 (1999)Google Scholar
  16. 16.
    Efros, A.A., Berg, A.C., Mori, G., Malik, J.: Recognizing action at a distance. In: Proc. ICCV (2003)Google Scholar
  17. 17.
    Schodl, A., Szeliski, R., Salesin, D.H., Essa, I.: Video textures. In: Proc. ACM SIGGRAPH, pp. 489–498 (2000)Google Scholar
  18. 18.
    Polana, R., Nelson, R.: Recognition of motion from temporal texture. In: Proc. CVPR, pp. 129–134 (1992)Google Scholar
  19. 19.
    Szummer, M.: Temporal texture modeling. Master’s thesis, MIT Media Lab, Cambridge MA (1995)Google Scholar
  20. 20.
    Fablet, R., Bouthemy, P.: Motion recognition using nonparametric image motion models. IEEE PAMI 25, 1619–1624 (2003)CrossRefGoogle Scholar
  21. 21.
    Doretto, G., Cremers, D., Favaro, P., Soatto, S.: Dynamic texture segmentation. In: Proc. ICCV, pp. 1236–1242 (2003)Google Scholar
  22. 22.
    Shumway, R.H., Stoffer, D.S.: Time Series Analysis and its Applications. Springer, Heidelberg (2000)CrossRefMATHGoogle Scholar
  23. 23.
    Manolakis, D.G., Ingle, V.K., Kogon, S.M.: Statistical and Adaptive Signal Processing. McGraw-Hill, Boston (2000)Google Scholar
  24. 24.
    Martin, R.: A metric for ARMA processes. IEEE Trans. on Signal Processing 48, 1164–1170 (2000)MathSciNetCrossRefMATHGoogle Scholar
  25. 25.
    Ripley, B.D.: Why do nearest-neighbour algorithms do so well? SIMCAT (Similarity and Categorization), Edinburgh (1997)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Franco Woolfe
    • 1
  • Andrew Fitzgibbon
    • 2
  1. 1.Yale UniversityNew HavenUSA
  2. 2.Microsoft ResearchCambridgeUK

Personalised recommendations