Video surveillance of large facilities, such as airports, rail stations and casinos, is developing rapidly. Cameras installed at such locations often overlook large crowds, which makes problems such as activity and scene understanding very challenging. Traditional activity recognition methods which rely on input from lower level processing units dealing with background subtraction and tracking, are unable to cope with frequent occlusions in such scenes. In this paper, we propose a novel activity representation and recognition method that bypasses these commonly used low level modules. We model each local spatio-temporal patch as a dynamic texture. Using a Martin distance metric to compare two patches based on their estimated dynamic texture parameters, we present a method to temporally stitch together local regions to form activity streamlines and represent each streamline by its constituent dynamic textures. This allows us to seamlessly perform activity recognition without explicitly detecting individuals in the scene. We demonstrate our method on multiple real data sets and show promising results.


Video Sequence Activity Representation Activity Recognition Local Binary Pattern Video Surveillance 
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.
    Veeraraghavan, A., RoyChowdhury, A.K., Chellappa, R.: Matching Shape Sequences in Video with Applications in Human Movement Analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence 27 (December 2005)Google Scholar
  2. 2.
    Ma, Y., Damelin, S.B., Masoud, O., Papanikolopoulos, N.: Activity Recognition Via Classification Constrained Diffusion Map. In: International Symposium on Visual Computing 2006, pp. 1–8 (2006)Google Scholar
  3. 3.
    Ma, Y., Miller, B., Buddharaju, P., Bazakos, M.: Activity Awareness: From Predefined Events to New Pattern Discovery. In: IEEE International Conference on Vision Systems, New York, NY, USA, January 5-7 (2006)Google Scholar
  4. 4.
    Park, S., Aggarwal, J.K.: A Hierarchical Bayesian Network for Event Recognition of Human Actions and Interactions. Multimedia Systems; Special issue on Video Surveillance 10(2), 164–179 (2004)Google Scholar
  5. 5.
    Muncaster, J., Ma, Y.: Hierarchical Model-Based Activity Recognition With Automatic Low-Level State Discovery. Journal of Multimedia 2(5), 66–76 (2007)CrossRefGoogle Scholar
  6. 6.
    Doretto, G., Chiuso, A., Wu, Y.N., Soatto, S.: Dynamic Textures. International Journal of Computer Vision 51(2), 91–109 (2003)CrossRefzbMATHGoogle Scholar
  7. 7.
    Saisan, P., Doretto, G., Wu, Y.N., Soatto, S.: Dynamic Texture Recognition. In: IEEE Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 58–63 (2001)Google Scholar
  8. 8.
    Chan, A., Vasconcelos, N.: Modeling, Clustering and Segmentating Video with Mixtures of Dynamic Textures. IEEE Trans. on Pattern Analysis and Machine Intelligence 30(5), 909–926 (2008)CrossRefGoogle Scholar
  9. 9.
    Fujita, K., Nayar, S.K.: Recognition of Dynamic Textures using Impulse Response of State Variables. In: 3rd International Workshop on Texture analysis and synthesis (2003)Google Scholar
  10. 10.
    Zhao, G., Pietikainen, M.: Dynamic Texture Recognition Using Volume Local Binary Patterns. In: 9th European Conference on Computer Vision, Graz, Austria, May 7 - 13 (2006)Google Scholar
  11. 11.
    Ma, Y., Cisar, P.: Motion Analysis Using Dynamic Texture in Crowd Environment. Image Analysis - From Theory to Applications, Research Publishing, pp. 49–54 (2008)Google Scholar
  12. 12.
    Ma, Y., Miller, B., Cohen, I.: Video Sequence Querying Using Clustering of Objects’ Appearance Models. In: Bebis, G., et al. (eds.) International Symposium on Visual Computing 2007, pp. 328–339 (2007)Google Scholar
  13. 13.
    i-lids dataset for avss (2007)Google Scholar
  14. 14.
    Ali, S., Shah, M.: A Lagrangian Particle Dynamic Approach for Crowd Flow Segmenta-tion and Stability Analysis. In: IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN 2007 (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Yunqian Ma
    • 1
  • Petr Cisar
    • 2
  1. 1.Honeywell LabsGolden ValleyUSA
  2. 2.Honeywell Prague LaboratoryPragueCzech Republic

Personalised recommendations