Title : Meta-Tracking for Video Scene Understanding

Speaker: Pierre-Marc Jodoin

Time: 27th Sep (Friday) 15:00

Location: C505

Abstract : this presentation is about a novel method to extract dominant motion patterns (MPs) and the main entry/exit areas from a surveillance video. The method use a novel particle meta-tracking procedure which produces meta-tracks, i.e. particle trajectories. As opposed to conventional tracking which focuses on individual moving objects, meta-tracking uses particles to follow the dominant flow of the traffic.

In a last step, a novel method is used to simultaneously identify the main entry/exit areas and recover the predominant MPs.  The meta-tracking procedure is a unique way to connect low-level motion features to long-range MPs. This kind of tracking is inspired by brain fiber tractography which has long been used to find dominant connections in the brain. Our method is fast, simple to implement, and works both on sparse and extremely crowded scenes. It also works on highly structured scenes (highways, traffic-light corners, etc.) as well as on chaotic scenes.

Bio:  Pierre-Marc Jodoin is an associate professor at the University of Sherbrooke in Canada.  He is a specialist in computer vision, image processing, video analytics, and medical imaging.  He co-founded in 2010 the Sherbrooke medical image processing service, co-founded in 2011 the compagny "imeka.ca" specialized in medical imaging and brain tractography and started the "changedetection.net" initiative in 2011.  He is currently the director of the Sherbrooke research center on intelligent environments which he also co-founded.  He is currently associate editor of IEEE transactions on image processing and guest editor of Pattern Recognition.