Multiple-person tracking system for content analysis

Jun-Wei Hsieh*, L. W. Huang, Yea Shuan Huang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

This paper presents a framework to track multiple persons in real-time. First, a method with real-time and adaptable capability is proposed to extract face-like regions based on skin, motion, and silhouette features. Then, a two-stage face verification algorithm is proposed to quickly eliminate false faces based on the face geometries and the Support Vector Machine(SVM). In order to overcome the effect of lighting changes, a method of color constancy compensation is applied. Then, a robust tracking scheme is applied to track multiple persons based on a face-status table. With the table, the system has extreme capabilities to track different persons at different statuses, which is quite important in face-related applications. Experimental results show that the proposed method is much robust and powerful than other traditional methods.

Original languageEnglish
Title of host publicationAdvances in Multimedia Information Processing - PCM 2001 - 2nd IEEE Pacific Rim Conference on Multimedia, Proceedings
EditorsShih-Fu Chang, Heung-Yeung Shum, Mark Liao
PublisherSpringer Verlag
Pages897-902
Number of pages6
ISBN (Print)3540426809, 9783540426806
DOIs
StatePublished - 1 Jan 2001
Event2nd IEEE Pacific-Rim Conference on Multimedia, IEEE-PCM 2001 - Beijing, China
Duration: 24 Oct 200126 Oct 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2195
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2nd IEEE Pacific-Rim Conference on Multimedia, IEEE-PCM 2001
CountryChina
CityBeijing
Period24/10/0126/10/01

Fingerprint Dive into the research topics of 'Multiple-person tracking system for content analysis'. Together they form a unique fingerprint.

Cite this