File Download
 
Links for fulltext
(May Require Subscription)
 
Supplementary

Conference Paper: Accurate foreground segmentation without pre-learning
  • Basic View
  • Metadata View
  • XML View
TitleAccurate foreground segmentation without pre-learning
 
AuthorsKuang, Z1
Zhou, H1
Wong, KKY1
 
KeywordsSegmentation
Contrast attenuation
Foreground segmentation
Graph cut
Automatic algorithms
 
Issue Date2011
 
PublisherIEEE Computer Society. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1001790
 
CitationThe 6th International Conference on Image and Graphics (ICIG 2011), Hefei, Anhui, China, 12-15 August 2011. In Proceedings of the 6th ICIG, 2011, p. 331-337 [How to Cite?]
DOI: http://dx.doi.org/10.1109/ICIG.2011.150
 
AbstractForeground segmentation has been widely used in many computer vision applications. However, most of the existing methods rely on a pre-learned motion or background model, which will increase the burden of users. In this paper, we present an automatic algorithm without pre-learning for segmenting foreground from background based on the fusion of motion, color and contrast information. Motion information is enhanced by a novel method called support edges diffusion (SED) , which is built upon a key observation that edges of the difference image of two adjacent frames only appear in moving regions in most of the cases. Contrasts in background are attenuated while those in foreground are enhanced using gradient of the previous frame and that of the temporal difference. Experiments on many video sequences demonstrate the effectiveness and accuracy of the proposed algorithm. The segmentation results are comparable to those obtained by other state-of-the-art methods that depend on a pre-learned background or a stereo setup. © 2011 IEEE.
 
ISBN978-0-7695-4541-7
 
DOIhttp://dx.doi.org/10.1109/ICIG.2011.150
 
ReferencesReferences in Scopus
 
DC FieldValue
dc.contributor.authorKuang, Z
 
dc.contributor.authorZhou, H
 
dc.contributor.authorWong, KKY
 
dc.date.accessioned2011-08-26T14:30:33Z
 
dc.date.available2011-08-26T14:30:33Z
 
dc.date.issued2011
 
dc.description.abstractForeground segmentation has been widely used in many computer vision applications. However, most of the existing methods rely on a pre-learned motion or background model, which will increase the burden of users. In this paper, we present an automatic algorithm without pre-learning for segmenting foreground from background based on the fusion of motion, color and contrast information. Motion information is enhanced by a novel method called support edges diffusion (SED) , which is built upon a key observation that edges of the difference image of two adjacent frames only appear in moving regions in most of the cases. Contrasts in background are attenuated while those in foreground are enhanced using gradient of the previous frame and that of the temporal difference. Experiments on many video sequences demonstrate the effectiveness and accuracy of the proposed algorithm. The segmentation results are comparable to those obtained by other state-of-the-art methods that depend on a pre-learned background or a stereo setup. © 2011 IEEE.
 
dc.description.naturepublished_or_final_version
 
dc.description.otherThe 6th International Conference on Image and Graphics (ICIG 2011), Hefei, Anhui, China, 12-15 August 2011. In Proceedings of the 6th ICIG, 2011, p. 331-337
 
dc.identifier.citationThe 6th International Conference on Image and Graphics (ICIG 2011), Hefei, Anhui, China, 12-15 August 2011. In Proceedings of the 6th ICIG, 2011, p. 331-337 [How to Cite?]
DOI: http://dx.doi.org/10.1109/ICIG.2011.150
 
dc.identifier.doihttp://dx.doi.org/10.1109/ICIG.2011.150
 
dc.identifier.epage337
 
dc.identifier.hkuros191682
 
dc.identifier.isbn978-0-7695-4541-7
 
dc.identifier.scopuseid_2-s2.0-80052995268
 
dc.identifier.spage331
 
dc.identifier.urihttp://hdl.handle.net/10722/137652
 
dc.languageeng
 
dc.publisherIEEE Computer Society. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1001790
 
dc.publisher.placeUnited States
 
dc.relation.ispartofInternational Conference on Image and Graphics Proceedings
 
dc.relation.referencesReferences in Scopus
 
dc.rightsInternational Conference on Image and Graphics Proceedings. Copyright © IEEE Computer Society.
 
dc.rights©2011 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
 
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License
 
dc.subjectSegmentation
 
dc.subjectContrast attenuation
 
dc.subjectForeground segmentation
 
dc.subjectGraph cut
 
dc.subjectAutomatic algorithms
 
dc.titleAccurate foreground segmentation without pre-learning
 
dc.typeConference_Paper
 
<?xml encoding="utf-8" version="1.0"?>
<item><contributor.author>Kuang, Z</contributor.author>
<contributor.author>Zhou, H</contributor.author>
<contributor.author>Wong, KKY</contributor.author>
<date.accessioned>2011-08-26T14:30:33Z</date.accessioned>
<date.available>2011-08-26T14:30:33Z</date.available>
<date.issued>2011</date.issued>
<identifier.citation>The 6th International Conference on Image and Graphics (ICIG 2011), Hefei, Anhui, China, 12-15 August 2011. In Proceedings of the 6th ICIG, 2011, p. 331-337</identifier.citation>
<identifier.isbn>978-0-7695-4541-7</identifier.isbn>
<identifier.uri>http://hdl.handle.net/10722/137652</identifier.uri>
<description.abstract>Foreground segmentation has been widely used in many computer vision applications. However, most of the existing methods rely on a pre-learned motion or background model, which will increase the burden of users. In this paper, we present an automatic algorithm without pre-learning for segmenting foreground from background based on the fusion of motion, color and contrast information. Motion information is enhanced by a novel method called support edges diffusion (SED) , which is built upon a key observation that edges of the difference image of two adjacent frames only appear in moving regions in most of the cases. Contrasts in background are attenuated while those in foreground are enhanced using gradient of the previous frame and that of the temporal difference. Experiments on many video sequences demonstrate the effectiveness and accuracy of the proposed algorithm. The segmentation results are comparable to those obtained by other state-of-the-art methods that depend on a pre-learned background or a stereo setup. &#169; 2011 IEEE.</description.abstract>
<language>eng</language>
<publisher>IEEE Computer Society. The Journal&apos;s web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1001790</publisher>
<relation.ispartof>International Conference on Image and Graphics Proceedings</relation.ispartof>
<rights>International Conference on Image and Graphics Proceedings. Copyright &#169; IEEE Computer Society.</rights>
<rights>&#169;2011 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.</rights>
<rights>Creative Commons: Attribution 3.0 Hong Kong License</rights>
<subject>Segmentation</subject>
<subject>Contrast attenuation</subject>
<subject>Foreground segmentation</subject>
<subject>Graph cut</subject>
<subject>Automatic algorithms</subject>
<title>Accurate foreground segmentation without pre-learning</title>
<type>Conference_Paper</type>
<description.nature>published_or_final_version</description.nature>
<identifier.doi>10.1109/ICIG.2011.150</identifier.doi>
<identifier.scopus>eid_2-s2.0-80052995268</identifier.scopus>
<identifier.hkuros>191682</identifier.hkuros>
<relation.references>http://www.scopus.com/mlt/select.url?eid=2-s2.0-80052995268&amp;selection=ref&amp;src=s&amp;origin=recordpage</relation.references>
<identifier.spage>331</identifier.spage>
<identifier.epage>337</identifier.epage>
<publisher.place>United States</publisher.place>
<description.other>The 6th International Conference on Image and Graphics (ICIG 2011), Hefei, Anhui, China, 12-15 August 2011. In Proceedings of the 6th ICIG, 2011, p. 331-337</description.other>
<bitstream.url>http://hub.hku.hk/bitstream/10722/137652/1/Content.pdf</bitstream.url>
</item>
Author Affiliations
  1. The University of Hong Kong