Xing HU, Shiqiang HU, Lingkun LUO, Guoxiang LI

Abnormal event detection in crowded scenes via bag-of-atomic-events-based topic model

: In this paper, we propose a novel framework for abnormal event detection in crowded scenes. A new concept of atomic event is introduced into this framework, which is the basic component of video events. Different from previous bag-of-words (BoW) modeling-based methods that represent feature descriptors using only one code word, a feature descriptor is represented using a few more atomic events in bag-of-atomic-events (BoAE) modeling. Consequently, the approximation error is reduced by using the obtained BoAE representation. In the context of abnormal event detection, BoAE representation is more suitable to describe abnormal events than BoW representation, because the abnormal event may not correspond to any code word in BoW modeling. Fast latent Dirichlet allocation is adopted to learn a model of normal events, as well as classify the testing event with low likelihood under the learned model. Our proposed framework is robust, computationally efficient, and highly accurate. We validate these advantages by conducting extensive experiments on several challenging datasets. Qualitative and quantitative results show the promising performance compared with other state-of-the-art methods.

PDF

___

[1] Krausz B, Bauckhage C. Loveparade 2010: Automatic video analysis of a crowd disaster. Comput Vis Image Und 2012; 116: 307-319.
[2] Stauffer C, Grimson WEL. Learning patterns of activity using real-time tracking. IEEE T Pattern Anal 2000; 22: 747-757.
[3] Piciarelli C, Micheloni C, Foresti GL. Trajectory-based anomalous event detection. IEEE T Circ Syst Vid 2008; 18: 1544-1554.
[4] Chen TP, Haussecker H, Bovyrin A, Belenov R, Rodyushkin K, Kuranov A, Eruhimov V. Computer vision workload analysis: case study of video surveillance systems. Intel Techn J 2005; 9: 109-118.
[5] Johnson N, Hogg D. Learning the distribution of object trajectories for event recognition. Image Vision Comput 1996; 14: 609-615.
[6] Jiang F, Wu Y, Katsaggelos AK. A dynamic hierarchical clustering method for trajectory-based unusual video event detection. IEEE T Image Process 2009; 18: 907-913.
[7] Zhang Y, Qin L, Yao H, Huang Q. Abnormal crowd behavior detection based on social attribute-aware force model. In: 19th IEEE International Conference on Image Processing; 30 September3 October 2012; Orlando, FL, USA. pp. 2689-2692.
[8] Zhang Y, Huang Q, Qin L, Zhao S, Yao H, Xu P. Dense crowd event recognition using bag of trajectory graphs. SIGNAL Image Video P 2014; 8 (Suppl. 1): S173-S181.
[9] Zhang Y, Zhang S, Huang Q, Thomas S. Learning sparse prototypes for crowd perception via ensemble coding mechanisms. In: 5th International Workshop on Human Behavior Understanding; 12 September 2014. pp 86-100.
[10] Zhang Y, Qin L, Ji R, Yao, H, Huang Q. Social attribute-aware force model: exploiting richness of interaction for abnormal crowd detection. IEEE T Circ Syst Vid 2014; 25: 1231-1245.
[11] Mehran R, Oyama A, Shah M. Abnormal crowd behavior detection using social force model. In: IEEE 2009 Computer Vision and Pattern Recognition; 1926 June 2009; Miami Beach, FL, USA. pp. 935-942.
[12] Kim J, Grauman K. Observe locally, infer globally: a space-time MRF for detecting abnormal activities with incremental updates. In: IEEE 2009 Computer Vision and Pattern Recognition; 19 26 June 2009; Miami Beach, FL, USA. pp. 2921-2928.
[13] Mahadevan V, Li W, Bhalodia V, Vasconcelos N. Anomaly detection in crowded scenes. In: IEEE 2010 Computer Vision and Pattern Recognition; 1318 June 2010; San Francisco, CA, USA. pp. 1975-1981.
[14] Kratz L, Nishino K. Anomaly detection in extremely crowded scenes using spatio-temporal motion pattern models. In: IEEE 2009 Computer Vision and Pattern Recognition; 1926 June 2009; Miami Beach, FL, USA: IEEE. pp. 1446-1453.
[15] Cui X, Liu Q, Gao M, Metaxas DN. Abnormal detection using interaction energy potentials. In: IEEE 2011 Computer Vision and Pattern Recognition; 2025 June 2011; Colorado Springs, CO, USA. pp. 3161-3167.
[16] Haidar Sharif M, Djeraba C. An entropy method for abnormal activities detection in video streams. Pattern Recogn 2012; 45: 2543-2561.
[17] Saligrama V, Konrad J, Jodoin P. Video anomaly identification. IEEE Signal Proc Mag 2010; 27: 18-33.
[18] Wang X, Ma X, Grimson WEL. Unsupervised activity perception in crowded and complicated scenes using hierarchical Bayesian models. IEEE T Pattern Anal 2009; 31: 539-555.
[19] Li J, Gong S, Xiang T. Learning behavioural context. Int J Comput Vision 2012; 97: 276-304.
[20] Zhao B, Fei-Fei L, Xing EP. Online detection of unusual events in videos via dynamic sparse coding. In: IEEE 2011 Computer Vision and Pattern Recognition; 2025 June 2011; Colorado Springs, CO, USA. pp. 3313-3320.
[21] Cong Y, Yuan J, Liu J. Abnormal event detection in crowded scenes using sparse representation. Pattern Recogn 2012; 46: 1851-1864.
[22] Mairal J, Bach F, Ponce J. Online dictionary learning for sparse coding. In: ACM 2009 International Conference on Machine Learning; 1418 June 2009; Montreal, Canada. pp. 689-696.
[23] Raghavendra R, Del Bue A, Cristani M. Optimizing interaction force for global anomaly detection in crowded scenes. In: IEEE 2011 International Conference on Computer Vision Workshop; 2025 June 2011; Colorado Springs, CO, USA. pp. 136-143.
[24] Xu J, Denman S, Fookes C, Sridharan S. Unusual scene detection using distributed behaviour model and sparse representation. In: 2012 IEEE Ninth International Conference on Advanced Video and Signal-Based Surveillance; 1821 September 2012; Beijing, China. pp. 48-53.
[25] Adam A, Rivlin E, Shimshoni I, Reinitz D. Robust real-time unusual event detection using multiple fixed-location monitors. IEEE T Pattern Anal 2008; 30: 555-560.
[26] Bertini M, Del Bimbo A, Seidenari L. Multi-scale and real-time non-parametric method for anomaly detection and localization. Comput Vis Image Und 2012; 116: 320-329.
[27] Chandola V, Banerjee A, Kumar V. Anomaly detection: a survey. ACM Comput Surv 2009; 41: 15.
[28] Castrodad A, Sapiro G. Sparse modeling of human actions from motion imagery. Int J Comput Vision 2012; 100: 1-15.
[29] Blei DM, Ng AY, Jordan MI. Latent Dirichlet allocation. J Mach Learn Res 2003; 3: 993-1022.
[30] Vretos N, Nikolaidis N, Pitas I. Video fingerprinting using latent Dirichlet allocation and facial images. Pattern Recogn 2012; 45: 2489-2498.
[31] Wang Y, Mori G. Human action recognition by semi-latent topic models. IEEE T Pattern Anal 2009; 31: 1762-1774.
[32] Wright J, Ma Y, Mairal J, Sapiro G, Huang TS, Yan S. Sparse representation for computer vision and pattern recognition. P IEEE 2010; 98: 1031-1044.
[33] Guha T, Ward RK. Learning sparse representations for human action recognition. IEEE T Pattern Anal 2012; 34: 1576-1588.
[34] Shan H, Banerjee A. Mixed-membership naive Bayes models. Data Min Knowl Disc 2011; 23: 1-62.
[35] Baraniuk RG, Wakin MB. Random projections of smooth manifolds. Found Comput Math 2009; 9: 51-77.
[36] Aharon M, Elad M, Bruckstein A. K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation. IEEE T Signal Proces 2006; 54: 4311-4322.
[37] Hoffman M, Bach FR, Blei DM. Online learning for latent Dirichlet allocation. In: Advances in Neural Information Processing Systems; 2010. pp. 856-864.