File Download
 
Links for fulltext
(May Require Subscription)
 
Supplementary

Conference Paper: Human face tracking system with multiple non-overlapping cameras
  • Basic View
  • Metadata View
  • XML View
TitleHuman face tracking system with multiple non-overlapping cameras
 
AuthorsHe, XC1
Yuk, SC1
Luo, T1
Chow, KP1
Wong, KYK1
Chung, RHY1
 
Issue Date2011
 
CitationProceedings - 3Rd International Conference On Data Mining And Intelligent Information Technology Applications, Icmia 2011, 2011, p. 176-179 [How to Cite?]
 
AbstractA novel human face tracking system with multiple cameras is proposed in this article. Unlike most surveillance approaches requiring overlapping field of views (FOVs) for tracking targets across multiple cameras, our proposed system utilizes face recognition as cue to build correspondence between cameras, and is therefore compatible with application with non-overlapping FOVs. Face recognition achieves great success in recent years, however it has an intrinsic limitation of low tolerance to face pose changes. In the proposed system, face reconstruction technology is adopted to transform non-frontal human face to a frontal one to substantially enhance the usability of face recognition, so that correspondence of humans across multiple trajectories can be reliably established. © 2011 AICIT.
 
ISBN978-898867848-0
 
ReferencesReferences in Scopus
 
DC FieldValue
dc.contributor.authorHe, XC
 
dc.contributor.authorYuk, SC
 
dc.contributor.authorLuo, T
 
dc.contributor.authorChow, KP
 
dc.contributor.authorWong, KYK
 
dc.contributor.authorChung, RHY
 
dc.date.accessioned2012-06-26T06:32:39Z
 
dc.date.available2012-06-26T06:32:39Z
 
dc.date.issued2011
 
dc.description.abstractA novel human face tracking system with multiple cameras is proposed in this article. Unlike most surveillance approaches requiring overlapping field of views (FOVs) for tracking targets across multiple cameras, our proposed system utilizes face recognition as cue to build correspondence between cameras, and is therefore compatible with application with non-overlapping FOVs. Face recognition achieves great success in recent years, however it has an intrinsic limitation of low tolerance to face pose changes. In the proposed system, face reconstruction technology is adopted to transform non-frontal human face to a frontal one to substantially enhance the usability of face recognition, so that correspondence of humans across multiple trajectories can be reliably established. © 2011 AICIT.
 
dc.description.naturelink_to_subscribed_fulltext
 
dc.identifier.citationProceedings - 3Rd International Conference On Data Mining And Intelligent Information Technology Applications, Icmia 2011, 2011, p. 176-179 [How to Cite?]
 
dc.identifier.epage179
 
dc.identifier.isbn978-898867848-0
 
dc.identifier.scopuseid_2-s2.0-84863395052
 
dc.identifier.spage176
 
dc.identifier.urihttp://hdl.handle.net/10722/152030
 
dc.languageeng
 
dc.relation.ispartofProceedings - 3rd International Conference on Data Mining and Intelligent Information Technology Applications, ICMIA 2011
 
dc.relation.referencesReferences in Scopus
 
dc.titleHuman face tracking system with multiple non-overlapping cameras
 
dc.typeConference_Paper
 
<?xml encoding="utf-8" version="1.0"?>
<item><contributor.author>He, XC</contributor.author>
<contributor.author>Yuk, SC</contributor.author>
<contributor.author>Luo, T</contributor.author>
<contributor.author>Chow, KP</contributor.author>
<contributor.author>Wong, KYK</contributor.author>
<contributor.author>Chung, RHY</contributor.author>
<date.accessioned>2012-06-26T06:32:39Z</date.accessioned>
<date.available>2012-06-26T06:32:39Z</date.available>
<date.issued>2011</date.issued>
<identifier.citation>Proceedings - 3Rd International Conference On Data Mining And Intelligent Information Technology Applications, Icmia 2011, 2011, p. 176-179</identifier.citation>
<identifier.isbn>978-898867848-0</identifier.isbn>
<identifier.uri>http://hdl.handle.net/10722/152030</identifier.uri>
<description.abstract>A novel human face tracking system with multiple cameras is proposed in this article. Unlike most surveillance approaches requiring overlapping field of views (FOVs) for tracking targets across multiple cameras, our proposed system utilizes face recognition as cue to build correspondence between cameras, and is therefore compatible with application with non-overlapping FOVs. Face recognition achieves great success in recent years, however it has an intrinsic limitation of low tolerance to face pose changes. In the proposed system, face reconstruction technology is adopted to transform non-frontal human face to a frontal one to substantially enhance the usability of face recognition, so that correspondence of humans across multiple trajectories can be reliably established. &#169; 2011 AICIT.</description.abstract>
<language>eng</language>
<relation.ispartof>Proceedings - 3rd International Conference on Data Mining and Intelligent Information Technology Applications, ICMIA 2011</relation.ispartof>
<title>Human face tracking system with multiple non-overlapping cameras</title>
<type>Conference_Paper</type>
<description.nature>link_to_subscribed_fulltext</description.nature>
<identifier.scopus>eid_2-s2.0-84863395052</identifier.scopus>
<relation.references>http://www.scopus.com/mlt/select.url?eid=2-s2.0-84855839214&amp;selection=ref&amp;src=s&amp;origin=recordpage</relation.references>
<identifier.spage>176</identifier.spage>
<identifier.epage>179</identifier.epage>
</item>
Author Affiliations
  1. The University of Hong Kong