Real-time active tracking with level sets

Warakorn Gulyanon, Claire Morand, Neil Robertson, Andrew Michael Wallace

Research output: Chapter in Book/Report/Conference proceedingConference contribution

99 Downloads (Pure)

Abstract

This paper presents a new real-time active visual tracker which improves standard mean shift tracking by using level sets to extract contours from the target. We use colour and the disparity map computed from a stereo cam- era pair which prove to be powerful features for tracking in an indoor surveillance scenario. To combine the fea- tures in the level sets process, we enhance Chen’s et al appearance model of [5] by using a probabilistic model determined via Expectation-Maximization (EM) cluster- ing. The level set result is used as the weighting kernel which improves the accuracy of the similarity measure- ment in the mean shift method. Finally a Kalman filter deals with complete occlusions.
Original languageEnglish
Title of host publicationProc. 4th IET Conf. Imaging for Crime Detection and Prevention
PublisherInstitution of Engineering and Technology
Number of pages6
Publication statusPublished - 3 Nov 2011

Fingerprint Dive into the research topics of 'Real-time active tracking with level sets'. Together they form a unique fingerprint.

Cite this