CSDL Home IEEE Transactions on Pattern Analysis & Machine Intelligence 2012 vol.34 Issue No.09 - Sept.
Issue No.09 - Sept. (2012 vol.34)
C. C. Loy , Sch. of Electr. Eng. & Comput. Sci., Queen Mary Univ. of London, London, UK
Tao Xiang , Sch. of Electr. Eng. & Comput. Sci., Queen Mary Univ. of London, London, UK
Shaogang Gong , Sch. of Electr. Eng. & Comput. Sci., Queen Mary Univ. of London, London, UK
Activity modeling and unusual event detection in a network of cameras is challenging, particularly when the camera views are not overlapped. We show that it is possible to detect unusual events in multiple disjoint cameras as context-incoherent patterns through incremental learning of time delayed dependencies between distributed local activities observed within and across camera views. Specifically, we model multicamera activities using a Time Delayed Probabilistic Graphical Model (TD-PGM) with different nodes representing activities in different decomposed regions from different views and the directed links between nodes encoding their time delayed dependencies. To deal with visual context changes, we formulate a novel incremental learning method for modeling time delayed dependencies that change over time. We validate the effectiveness of the proposed approach using a synthetic data set and videos captured from a camera network installed at a busy underground station.
video surveillance, computer graphics, delays, image sensors, learning (artificial intelligence), probability, video cameras, underground station, incremental activity modeling, multiple disjoint camera views, unusual event detection, camera networks, context-incoherent patterns, time delayed dependencies, distributed local activities, multicamera activity model, time delayed probabilistic graphical model, TD-PGM, decomposed regions, directed links, visual context changes, incremental learning method, synthetic data set, video capturing, Cameras, Delay effects, Visualization, Context, Videos, Complexity theory, Member and Geographic Activities, incremental structure learning., Unusual event detection, multicamera activity modeling, time delay estimation
C. C. Loy, Tao Xiang, Shaogang Gong, "Incremental Activity Modeling in Multiple Disjoint Cameras", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.34, no. 9, pp. 1799-1813, Sept. 2012, doi:10.1109/TPAMI.2011.246