File Download
 
Links for fulltext
(May Require Subscription)
 
Supplementary

Conference Paper: Learning image-specific parameters for interactive segmentation
  • Basic View
  • Metadata View
  • XML View
TitleLearning image-specific parameters for interactive segmentation
 
AuthorsKuang, Z1
Schnieders, D1
Zhou, H1
Wong, KKY1
Yu, Y1
Peng, B2
 
KeywordsApproximation scheme
Conditional random field
Cutting plane methods
Energy margin
Interactive image segmentation
 
Issue Date2012
 
PublisherIEEE Computer Society. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000147
 
CitationThe IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI., 16-21 June 2012. In IEEE Conference on Computer Vision and Pattern Recognition Proceedings, 2012, p. 590-597 [How to Cite?]
DOI: http://dx.doi.org/10.1109/CVPR.2012.6247725
 
AbstractIn this paper, we present a novel interactive image segmentation technique that automatically learns segmentation parameters tailored for each and every image. Unlike existing work, our method does not require any offline parameter tuning or training stage, and is capable of determining image-specific parameters according to some simple user interactions with the target image. We formulate the segmentation problem as an inference of a conditional random field (CRF) over a segmentation mask and the target image, and parametrize this CRF by different weights (e.g., color, texture and smoothing). The weight parameters are learned via an energy margin maximization, which is solved using a constraint approximation scheme and the cutting plane method. Experimental results show that our method, by learning image-specific parameters automatically, outperforms other state-of-the-art interactive image segmentation techniques. © 2012 IEEE.
 
DescriptionPosters 1B - Color and Texture, Early & Biological Vision, Image Based Modeling, Segmentation and Grouping
 
ISBN978-1-4673-1228-8
 
ISSN1063-6919
2013 SCImago Journal Rankings: 2.954
 
DOIhttp://dx.doi.org/10.1109/CVPR.2012.6247725
 
DC FieldValue
dc.contributor.authorKuang, Z
 
dc.contributor.authorSchnieders, D
 
dc.contributor.authorZhou, H
 
dc.contributor.authorWong, KKY
 
dc.contributor.authorYu, Y
 
dc.contributor.authorPeng, B
 
dc.date.accessioned2012-07-16T09:53:34Z
 
dc.date.available2012-07-16T09:53:34Z
 
dc.date.issued2012
 
dc.description.abstractIn this paper, we present a novel interactive image segmentation technique that automatically learns segmentation parameters tailored for each and every image. Unlike existing work, our method does not require any offline parameter tuning or training stage, and is capable of determining image-specific parameters according to some simple user interactions with the target image. We formulate the segmentation problem as an inference of a conditional random field (CRF) over a segmentation mask and the target image, and parametrize this CRF by different weights (e.g., color, texture and smoothing). The weight parameters are learned via an energy margin maximization, which is solved using a constraint approximation scheme and the cutting plane method. Experimental results show that our method, by learning image-specific parameters automatically, outperforms other state-of-the-art interactive image segmentation techniques. © 2012 IEEE.
 
dc.description.naturepublished_or_final_version
 
dc.descriptionPosters 1B - Color and Texture, Early & Biological Vision, Image Based Modeling, Segmentation and Grouping
 
dc.description.otherThe IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI., 16-21 June 2012. In IEEE Conference on Computer Vision and Pattern Recognition Proceedings, 2012, p. 590-597
 
dc.identifier.citationThe IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI., 16-21 June 2012. In IEEE Conference on Computer Vision and Pattern Recognition Proceedings, 2012, p. 590-597 [How to Cite?]
DOI: http://dx.doi.org/10.1109/CVPR.2012.6247725
 
dc.identifier.doihttp://dx.doi.org/10.1109/CVPR.2012.6247725
 
dc.identifier.epage597
 
dc.identifier.hkuros200763
 
dc.identifier.isbn978-1-4673-1228-8
 
dc.identifier.issn1063-6919
2013 SCImago Journal Rankings: 2.954
 
dc.identifier.scopuseid_2-s2.0-84866689544
 
dc.identifier.spage590
 
dc.identifier.urihttp://hdl.handle.net/10722/152969
 
dc.languageeng
 
dc.publisherIEEE Computer Society. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000147
 
dc.publisher.placeUnited States
 
dc.relation.ispartofIEEE Conference on Computer Vision and Pattern Recognition Proceedings
 
dc.rightsIEEE Conference on Computer Vision and Pattern Recognition. Proceedings. Copyright © IEEE Computer Society.
 
dc.rights©2012 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.subjectApproximation scheme
 
dc.subjectConditional random field
 
dc.subjectCutting plane methods
 
dc.subjectEnergy margin
 
dc.subjectInteractive image segmentation
 
dc.titleLearning image-specific parameters for interactive segmentation
 
dc.typeConference_Paper
 
<?xml encoding="utf-8" version="1.0"?>
<item><contributor.author>Kuang, Z</contributor.author>
<contributor.author>Schnieders, D</contributor.author>
<contributor.author>Zhou, H</contributor.author>
<contributor.author>Wong, KKY</contributor.author>
<contributor.author>Yu, Y</contributor.author>
<contributor.author>Peng, B</contributor.author>
<date.accessioned>2012-07-16T09:53:34Z</date.accessioned>
<date.available>2012-07-16T09:53:34Z</date.available>
<date.issued>2012</date.issued>
<identifier.citation>The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI., 16-21 June 2012. In IEEE Conference on Computer Vision and Pattern Recognition Proceedings, 2012, p. 590-597</identifier.citation>
<identifier.isbn>978-1-4673-1228-8</identifier.isbn>
<identifier.issn>1063-6919</identifier.issn>
<identifier.uri>http://hdl.handle.net/10722/152969</identifier.uri>
<description>Posters 1B - Color and Texture, Early &amp; Biological Vision, Image Based Modeling, Segmentation and Grouping</description>
<description.abstract>In this paper, we present a novel interactive image segmentation technique that automatically learns segmentation parameters tailored for each and every image. Unlike existing work, our method does not require any offline parameter tuning or training stage, and is capable of determining image-specific parameters according to some simple user interactions with the target image. We formulate the segmentation problem as an inference of a conditional random field (CRF) over a segmentation mask and the target image, and parametrize this CRF by different weights (e.g., color, texture and smoothing). The weight parameters are learned via an energy margin maximization, which is solved using a constraint approximation scheme and the cutting plane method. Experimental results show that our method, by learning image-specific parameters automatically, outperforms other state-of-the-art interactive image segmentation techniques. &#169; 2012 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=1000147</publisher>
<relation.ispartof>IEEE Conference on Computer Vision and Pattern Recognition Proceedings</relation.ispartof>
<rights>IEEE Conference on Computer Vision and Pattern Recognition. Proceedings. Copyright &#169; IEEE Computer Society.</rights>
<rights>&#169;2012 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>Approximation scheme</subject>
<subject>Conditional random field</subject>
<subject>Cutting plane methods</subject>
<subject>Energy margin</subject>
<subject>Interactive image segmentation</subject>
<title>Learning image-specific parameters for interactive segmentation</title>
<type>Conference_Paper</type>
<description.nature>published_or_final_version</description.nature>
<identifier.doi>10.1109/CVPR.2012.6247725</identifier.doi>
<identifier.scopus>eid_2-s2.0-84866689544</identifier.scopus>
<identifier.hkuros>200763</identifier.hkuros>
<identifier.spage>590</identifier.spage>
<identifier.epage>597</identifier.epage>
<publisher.place>United States</publisher.place>
<description.other>The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Providence, RI., 16-21 June 2012. In IEEE Conference on Computer Vision and Pattern Recognition Proceedings, 2012, p. 590-597</description.other>
<bitstream.url>http://hub.hku.hk/bitstream/10722/152969/1/Content.pdf</bitstream.url>
</item>
Author Affiliations
  1. The University of Hong Kong
  2. Hong Kong Polytechnic University