An elliptical head tracker

Author

Birchfield, Stan

Author_Institution

Dept. of Comput. Sci., Stanford Univ., CA, USA

Volume

2

fYear

1997

fDate

2-5 Nov. 1997

Firstpage

1710

Abstract

A simple algorithm for tracking a person´s head is presented. A two-dimensional model, namely an ellipse, is used to approximate the head´s contour. When a new image becomes available, a local search determines the position and size of the best ellipse by maximizing the normalized sum of the image gradient magnitude around the perimeter of the ellipse. The local search begins from a predicted position, using the head´s velocity which eliminates the tracker´s dependence upon maximum velocity. The tracker operates at 30 Hz and actively controls camera pan and tilt in order to track a person moving in a real environment. The algorithm tolerates full 360-degree rotation of the body as well as moderate amounts of occlusion, and it performs reacquisition of the subject.

Keywords

edge detection; optical tracking; search problems; camera pan; ellipse; elliptical head tracker; head contour; image gradient magnitude; local search; occlusion; position; predicted position; reacquisition; size; tilt; two-dimensional model; velocity; Application software; Automatic control; Cameras; Computer aided instruction; Computer science; Computer vision; Head; Surveillance; Torso; Videoconference;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems & Computers, 1997. Conference Record of the Thirty-First Asilomar Conference on

Conference_Location

Pacific Grove, CA, USA

ISSN

1058-6393

Print_ISBN

0-8186-8316-3

Type

conf

DOI

10.1109/ACSSC.1997.679194

Filename

679194