offres d'emploi formations actualités contact accès annuaire intranet
Séminaires >

Automated Video-Based Fall Detection from Overlapping Cameras

Mikael A. Mousse, doctorant LISIC

jeudi 9 juillet 2015 à 14h15

B014


Automatically detecting falls is a desired part of caring for a live-alone senior. Automatically detecting a fall can enable rapid response that in turn can reduce additional complications from a long period in a fallen position. Even when a fall did not result in injury, automatic detection can alert caregivers of the need for preemptive measures such as hazard elimination, physical conditioning, etc. A fall is characterized by a person beginning with normal behaviors, such as sitting or walking, followed by the person rapidly descending then laying on the ground for an extended period of time. Researchers have developed various video-based fall detection methods, including moving-region-based 3D-projection-based methods. We investigate a video-based fall detection method from multiple cameras that will be simpler and more efficient than previous methods, while being equally or more accurate. In this talk, we present our contribution for moving object detection. We also present an overview of our fall detection method.