Mobile based Pedestrian Detection with Accurate Tracking

This paper presents an innovative smartphone application which takes advantage of the availability of smartphone technologies to develop a highly accurate pedestrian detection and tracking application, based on monocular camera and embedded sensors. Monocular pedestrian detection is performed based on well-known vision approaches, later pose estimation is used to correct the location of the pedestrian, based on the information provided by the mobile device, and finally movement is estimated by means of a Kalman Filter. The information from the internal sensors corrects the distance given by the pin-hole model. Test performed proved the viability of the system as well as its accuracy.