Skip to Main content Skip to Navigation
New interface
Conference papers

Multi-Camera Visual Surveillance for Motion Detection, Occlusion Handling, Tracking and Event Recognition

Abstract : This paper presents novel approaches for background modeling, occlusion handling and event recognition by using multi-camera configurations that can be used to overcome the limitations of the single camera configurations. The main novelty in proposed background modeling approach is building multivariate Gaussians background model for each pixel of the reference camera by utilizing homography-related positions. Also, occlusion handling is achieved by generation of the top-view via trifocal tensors, as a result of matching over-segmented regions instead of pixels. The resulting graph is segmented into objects after determining the minimum spanning tree of this graph. Tracking of multi-view data is obtained by utilizing measurements across the views in case of occlusions. The last contribution is the classification of the resulting trajectories by GM-HMMs, yielding better results for using together all different view trajectories of the same object. Hence, multi-camera sensing is fully exploited from motion detection to event modeling.
Document type :
Conference papers
Complete list of metadata

Cited literature [16 references]  Display  Hide  Download
Contributor : Peter Sturm Connect in order to contact the contributor
Submitted on : Sunday, October 5, 2008 - 3:06:03 PM
Last modification on : Monday, October 6, 2008 - 9:17:15 AM
Long-term archiving on: : Monday, October 8, 2012 - 2:01:01 PM


Files produced by the author(s)


  • HAL Id : inria-00326780, version 1



Oytun Akman, A. Aydin Alatan, Tolga Çiloglu. Multi-Camera Visual Surveillance for Motion Detection, Occlusion Handling, Tracking and Event Recognition. Workshop on Multi-camera and Multi-modal Sensor Fusion Algorithms and Applications - M2SFA2 2008, Andrea Cavallaro and Hamid Aghajan, Oct 2008, Marseille, France. ⟨inria-00326780⟩



Record views


Files downloads