Advertisement

Scene Segmentation for Behaviour Correlation

  • Jian Li
  • Shaogang Gong
  • Tao Xiang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5305)

Abstract

This paper presents a novel framework for detecting abnormal pedestrian and vehicle behaviour by modelling cross-correlation among different co-occurring objects both locally and globally in a given scene. We address this problem by first segmenting a scene into semantic regions according to how object events occur globally in the scene, and second modelling concurrent correlations among regional object events both locally (within the same region) and globally (across different regions). Instead of tracking objects, the model represents behaviour based on classification of atomic video events, designed to be more suitable for analysing crowded scenes. The proposed system works in an unsupervised manner throughout using automatic model order selection to estimate its parameters given video data of a scene for a brief training period. We demonstrate the effectiveness of this system with experiments on public road traffic data.

Keywords

False Alarm Video Clip Anomaly Detection Spectral Cluster Image Event 
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.

References

  1. 1.
    Hu, W., Xiao, X., Fu, Z., Xie, D., Tan, T., Maybank, S.: A system for learning statistical motion patterns. PAMI 28 (9), 1450–1464 (2006)Google Scholar
  2. 2.
    Johnson, N., Hogg, D.: Learning the distribution of object trajectories for event recognition. BMVC 2, 583–592 (1995)Google Scholar
  3. 3.
    Xiang, T., Gong, S.: Beyond tracking: Modelling activity and understanding behaviour. IJCV 67 (1), 21–51 (2006)CrossRefGoogle Scholar
  4. 4.
    Xiang, T., Gong, S.: Video behavior profiling for anomaly detection. PAMI 30(5), 893–908 (2008)Google Scholar
  5. 5.
    Brand, M., Kettnaker, V.: Discovery and segmentation of activities in video. PAMI 22(8), 844–851 (2000)Google Scholar
  6. 6.
    Wang, X., Ma, X., Grimson, W.E.L.: Unsupervised activity perception by hierarchical bayesian models. In: CVPR, Minneapolis, USA, June 18-23, pp. 1–8 (2007)Google Scholar
  7. 7.
    Stauffer, C., Grimson, W.E.L.: Adaptive background mixture models for real-time tracking. In: CVPR, vol. 2, pp. 246–252 (1999)Google Scholar
  8. 8.
    Russell, D., Gong, S.: Minimum cuts of a time-varying background. In: BMVC, Edinburgh, UK, 1–10 (September 2006)Google Scholar
  9. 9.
    Gong, S., Xiang, T.: Scene event recognition without tracking. Special issue on visual surveillance, Acta Automatica Sinica 29(3), 321–331 (2003)Google Scholar
  10. 10.
    Schwarz, G.: Estimating the dimension of a model. Annals of Statistics 6(2), 461–464 (1978)zbMATHCrossRefMathSciNetGoogle Scholar
  11. 11.
    Dempster, A., Laird, N., Rubin, D.: Maximum likelihood from incomplete data via the em algorithm. Journal of the Royal Statistical Society, series B 39(1), 1–38 (1977)zbMATHMathSciNetGoogle Scholar
  12. 12.
    Zelnik-Manor, L., Perona, P.: Self-tuning spectral clustering. In: NIPS (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Jian Li
    • 1
  • Shaogang Gong
    • 1
  • Tao Xiang
    • 1
  1. 1.Department of Computer Science Queen Mary CollegeUniversity of LondonLondonUK

Personalised recommendations