CONF Naturel_ICPR_2008/IDIAP Detecting queues at vending machines: a statistical layered approach Naturel, Xavier Odobez, Jean-Marc EXTERNAL http://publications.idiap.ch/attachments/papers/2008/Naturel_ICPR_2008.pdf PUBLIC http://publications.idiap.ch/index.php/publications/showcite/naturel:rr08-04 Related documents Proc. Int. Conf. on Pattern Recognition (ICPR) Tampa 2008 December 2008 This paper presents a method for monitoring activities at a ticket vending machine in a video-surveillance context. Rather than relying on the output of a tracking module, which is prone to errors, the events are direclty recognized from image measurements. This especially does not require tracking. A statistical layered approach is proposed, where in the first layer, several sub-events are defined and detected using a discriminative approach. The second layer uses the result of the first and models the temporal relationships of the high-level event using a Hidden Markov Model (HMM). Results are assessed on 3h30 hours of real video footage coming from Turin metro station. REPORT naturel:rr08-04/IDIAP Detecting queues at vending machines: a statistical layered approach Naturel, Xavier Odobez, Jean-Marc EXTERNAL http://publications.idiap.ch/attachments/reports/2008/naturel-idiap-rr-08-04.pdf PUBLIC Idiap-RR-04-2008 2008 IDIAP In this report, a method for monitoring activity at a ticket machine is presented. While this work has been done in the specific context of Turin metro, the proposed model could be applied to other locations and tasks in video-surveillance. Monitoring the activity is based here on event recognition, by modelling directly the events of interest.We especially focus on detecting queues at ticket vending machines. A 2-layer model is proposed. In the first layer, several sub-events are defined and detected using a discriminative approach (SVMs). The second layer uses the result of the first and model the high-level event of interest. Results are assessed on 4 hours of real video footage coming from Turin metro station.