MIMOS Research Seminar: Multi-camera Anomaly Detection

WHAT : Multi-camera Anomaly Detection
WHO: Chen Change, LOY (Queens Mary College, University of London, U.K.)
WHEN: 1100 am – 1200 noon ,  Tuesday 22 December 2009
WHERE: MIMOS Berhad, Technology Park Malaysia, 57000 Kuala Lumpur

Admission : Free, but registration is encouraged..
Parking : There is ample free parking within the grounds of MIMOS Bhd.

Abstract :
We present a novel approach for detecting global behaviour anomalies in multiple disjoint cameras by learning time delayed dependencies between activities cross camera views. Specifically, we propose to model multi-camera activities using a Time Delayed Probabilistic Graphical Model (TD-PGM) with different nodes representing activities in different semantically decomposed regions from different camera views, and the directed links between nodes encoding causal relationships between the activities. A novel two-stage structure learning algorithm is formulated to learn globally optimised time-delayed dependencies. A new cumulative abnormality score is also introduced to replace the conventional log-likelihood score for gaining significantly more robust and reliable real-time anomaly detection. The effectiveness of the proposed approach is validated using a camera network installed at a busy underground station.

Profile :
Loy Chen Change is a PhD candidate in Queen Mary, University of London, U.K., under the supervision of Dr. Xiang  and Prof. Gong. Previous to this, he was a researcher in Centre of Multimodel Signal Processing, Mimos Berhad. His work is focused on computer vision, pattern recognition and image processing. He has published in numerous prestigious computer vision conferences, e.g. ICCV, CVPR and BMVC.

All are welcome!  However, as there are only a limited number of seats available, please come early to avoid the disappointment of not having a place to sit for the entire duration of the presentation.

We would appreciate receiving confirmation of participation (email: cs.chan@mimos.my or tel. 03 8995 5000 ext. 6270). See you there !