Traffic Pattern Analysis and Anomaly Detection via Probabilistic Inference Model

Part of the KAIST Research Series book series (KAISTRS)


In this chapter, we introduce a method for trajectory pattern analysis through the probabilistic inference model with both regional and velocity observations. By embedding Gaussian models into the discrete topic model framework, our method uses continuous velocity as well as regional observations unlike the existing approaches. In addition, the proposed framework combined with Hidden Markov Model can cover the temporal transition of the scene state, which is useful in checking violation of the rule that some conflict topics (e.g., two cross traffic patterns) should not occur at the same time. To achieve online learning even with the complexity of the proposed model, we suggest a novel learning scheme instead of collapsed Gibbs sampling. The proposed two-stage greedy learning scheme is not only efficient at reducing the search space but also accurate in a way that the accuracy of online learning becomes not worse than that of the batch learning. To validate the performance of our method, experiments were conducted on various datasets. Experimental results show that our model explains satisfactorily the trajectory patterns with respect to scene understanding, anomaly detection, and prediction.


Trajectory pattern analysis Anomaly detection in traffic Probabilistic inference model Topic model Online inference learning Scene understanding 



This work was sponsored by Samsung Techwin Co.,Ltd and BK 21 plus program, and also partially supported by the Center for Integrated Smart Sensors funded by the Ministry of Science, ICT & Future Planning as the Global Frontier Project.


  1. 1.
    Basharat A, Gritai A, Shah M (2008) Learning object motion patterns for anomaly detection and improved object detection. In: IEEE Conference on CVPRGoogle Scholar
  2. 2.
    Benezeth Y, Jodoin PM, Saligrama V (2011) Abnormality detection using low-level co-occurring events. Pattern Recognit Lett 32(3):423–431CrossRefGoogle Scholar
  3. 3.
    Bishop CM (2006) Pattern recognition and machine learning (Information science and statistics). Springer-Verlag New York Inc, SecaucusGoogle Scholar
  4. 4.
    Blei DM, Ng AY, Jordan MI (2003) Latent dirichlet allocation. JML. Res. 3:993–1022zbMATHGoogle Scholar
  5. 5.
    Canini KR, Shi L, Griffiths TL (2009) Online inference of topics with latent dirichlet allocation. In: AI-STATSGoogle Scholar
  6. 6.
    Duda RO, Hart PE, Stork DG (2000) Pattern classification, 2nd edn. Wiley-Interscience, New YorkGoogle Scholar
  7. 7.
    Emonet R, Varadarajan J, Odobez JM (2011) Extracting and locating temporal motifs in video scenes using a hierarchical non parametric bayesian model. In: IEEE conference on CVPR, pp 3233–3240Google Scholar
  8. 8.
    Griffiths TL, Steyvers M (2004) Finding scientific topics. PNAS 101(Suppl 1):5228–5235CrossRefGoogle Scholar
  9. 9.
    Hoffman M, Blei DM, Bach F (2010) Online learning for latent dirichlet allocation. In: NIPSGoogle Scholar
  10. 10.
    Hospedales TM, Gong S, Xiang T (2009) A markov clustering topic model for mining behaviour in video. In: ICCV, pp 1165–1172. IEEEGoogle Scholar
  11. 11.
    Hu W, Xiao X, Fu Z, Xie D, Tan T, Maybank S (2006) A system for learning statistical motion patterns. IEEE Trans Pattern Anal Mach Intell 28(9):1450–1464CrossRefGoogle Scholar
  12. 12.
    Jeong H, Yoo YJ, Yi KM, Choi JY (2014) Two-stage online inference model for traffic pattern analysis and anomaly detection. Mach Vis Appl 25(6):1501–1517CrossRefGoogle Scholar
  13. 13.
    Kratz L, Nishino K (2009) Anomaly detection in extremely crowded scenes using spatio-temporal motion pattern models. In: 2013 IEEE conference on computer vision and pattern recognition 0, 1446–1453Google Scholar
  14. 14.
    Kuette D, Breitenstein MD, Van Gool L, Ferrari V (2010) What’s going on? Discovering spatio-temporal dependencies in dynamic scenes. In: CVPR, pp 1951–1958. doi: 10.1109/CVPR.2010.5539869
  15. 15.
    Machy C, Desurmont X, Delaigle JF, Bastide A (2007) Introduction of cctv at level crossings with automatic detection of potentially dangerous situationsGoogle Scholar
  16. 16.
    Mahadevan V, Li W, Bhalodia V, Vasconcelos N (2010) Anomaly detection in crowded scenes. In: IEEE conference on CVPR, pp 1975–1981Google Scholar
  17. 17.
    Morris B, Trivedi M (2008) A survey of vision-based trajectory learning and analysis for surveillance. IEEE Trans Circuits Syst Video Technol 18(8):1114–1127CrossRefGoogle Scholar
  18. 18.
    Morris B, Trivedi MM (2009) Learning trajectory patterns by clustering: Experimental studies and comparative evaluation. In: CVPR, pp 312–319Google Scholar
  19. 19.
    Piciarelli C, Foresti GL (2006) Online trajectory clustering for anomalous events detection. Pattern Recognit Lett 1835–1842Google Scholar
  20. 20.
    Qin Z, Shelton CR (2012) Improving multi-target tracking via social grouping. In: IEEE conference on computer vision and pattern recognitionGoogle Scholar
  21. 21.
    Rodriguez M, Ali S, Kanade T (2009) Tracking in unstructured crowded scenes. In: ICCV, pp 1389–1396. IEEEGoogle Scholar
  22. 22.
    Saleemi I, Hartung L, Shah M (2010) Scene understanding by statistical modeling of motion patterns. In: CVPR, pp 2069–2076. IEEEGoogle Scholar
  23. 23.
    Saleemi I, Shafique K, Shah M (2009) Probabilistic modeling of scene dynamics for applications in visual surveillance. IEEE Trans PAMI 31(8):1472–1485CrossRefzbMATHGoogle Scholar
  24. 24.
    Stauffer C, Grimson WEL (1999) Adaptive background mixture models for real-time tracking. In: CVPR, pp 2246–2252Google Scholar
  25. 25.
    Tomasi C, Kanade T (1991) Detection and tracking of point features. Techical report, IJCVGoogle Scholar
  26. 26.
  27. 27.
    UMN: Crowd dataset.
  28. 28.
    Varadarajan J, Emonet R, Odobez J (2012) Bridging the past, present and future: Modeling scene activities from event relationships and global rules. In: IEEE conference on CVPR, pp 2096–2103Google Scholar
  29. 29.
    Walk S, Majer N, Schindler K, Schiele B (2010) New features and insights for pedestrian detection. In: Conference on CVPR. IEEE, San FranciscoGoogle Scholar
  30. 30.
    Wang B, Ye M, Li X, Zhao F, Ding J (2012) Abnormal crowd behavior detection using high-frequency and spatio-temporal features. Mach Vis Appl 23(3):501–511CrossRefzbMATHGoogle Scholar
  31. 31.
    Wang X, Ma KT, Ng GW, Grimson WE (2011) Trajectory analysis and semantic region modeling using nonparametric hierarchical bayesian models. Int J Comput Vis 95(3):287–312CrossRefGoogle Scholar
  32. 32.
    Wang X, Ma X, Grimson WEL (2009) Unsupervised activity perception in crowded and complicated scenes using hierarchical bayesian models. IEEE Trans PAMI 31(3):539–555CrossRefGoogle Scholar
  33. 33.
    Wang X, Tieu K, Grimson E (2006) Learning semantic scene models by trajectory analysis. In: Proceedings of the 9th ECCV, vol Part III, ECCV’06. Springer, Berlin, pp 110–123Google Scholar
  34. 34.
    Zhai K, Boyd-Graber J, Asadi N, Alkhouja M (2012) Mr. LDA: A flexible large scale topic modeling package using variational inference in mapreduce. In: ACM International conference on world wide webGoogle Scholar
  35. 35.
    Zhao B, Fei-Fei L, Xing EP (2011) Online detection of unusual events in videos via dynamic sparse coding. In: IEEE conference on CVPR. Colorado Springs, COGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2016

Authors and Affiliations

  1. 1.Perception and Intelligence Laboratory, ASRI Room 413, Bldg 133Seoul National UniversityGwanak-guKorea
  2. 2.Ecole Polytechnique Federale de Lausanne, EPFL, CVLABLausanneSwitzerland

Personalised recommendations