Model-Based Object Tracking in Traffic Scenes

D. Koller, K. Daniilidis, T. Thorhallson, and H.-H. Nagel

In Proc. Second European Conference on Computer Vision, pp. 437-452, S. Margherita, Ligure, Italy, May 18-23, 1992, LNCS 588, Springer-Verlag, 1992.

Abstract

This contribution addresses the problem of detection and tracking of moving vehicles in image sequences from traffic scenes recorded by a stationary camera. In order to exploit the a priori knowledge about the shape and the physical motion of vehicles in traffic scenes, a parameterized vehicle model is used for an intraframe matching process and a recursive estimator based on a motion model is used for motion estimation. The initial guess about the position and orientation for the models are computed with the help of a clustering approach of moving image features. Shadow edges of the models are taken into account in the matching process. This enables tracking of vehicles under complex illumination conditions and within a small effective field of view. Results on real world traffic scenes are presented and open problems are outlined.

For a printed copy send email to koller@vision.caltech.edu


Last modified on Tuesday, November 20, 1996, Dieter Koller (koller@vision.caltech.edu)