Conference Paper: Sub-scene generation: a step towards complex scene understanding

File Download

Content.pdf

Links for fulltext
(May Require Subscription)

Publisher Website: 10.1109/ICME.2011.6011910
Scopus: eid_2-s2.0-80155132496

Supplementary

Citations:
- Scopus:
- Web of Science:
- PubMed Central:
Item Statistics (The Hub)
Appears in Collections:
- Electrical & Electronic Engineering: Conference papers

Title	Sub-scene generation: a step towards complex scene understanding
Authors	Zhu, S Yung, NHC1
Keywords	Sub-scene Semantic gap Complex scene Color harmony Image patches
Issue Date	2011
Publisher	IEEE. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000477
Citation	The 2011 IEEE International Conference on Multimedia and Expo (ICME 2011), Barcelona, Spain, 11-15 July 2011. In Conference Proceedings of ICME, 2011, p. 1-6 [How to Cite?] DOI: http://dx.doi.org/10.1109/ICME.2011.6011910
Abstract	This paper proposes a new method for generating sub-scenes from images of natural scenes. It is based on over-segmented image patches initially and uses rules derived from human psychology to merge the patches into reasonable sub-scenes. This merging process focuses on covering semantic gaps by exploring semantic connections among the patches, such as the influence area of each patch where the potential interaction with other patches may exist; and applying visual organization rules of proximity property and color harmony property. The proposed method is simple yet efficient computationally. It does not have any limitation on the input image class and does not include a training process. Our experiments indicate that it handles complex scenes with clutters and occlusions very well, and outperforms general segmentation methods by its meaningful semantic grouping and self determined stop criteria. © 2011 IEEE.
ISBN	978-1-61284-350-6
DOI	http://dx.doi.org/10.1109/ICME.2011.6011910
References	References in Scopus

DC Field	Value
dc.contributor.author	Zhu, S
dc.contributor.author	Yung, NHC
dc.date.accessioned	2012-08-08T09:01:05Z
dc.date.available	2012-08-08T09:01:05Z
dc.date.issued	2011
dc.description.abstract	This paper proposes a new method for generating sub-scenes from images of natural scenes. It is based on over-segmented image patches initially and uses rules derived from human psychology to merge the patches into reasonable sub-scenes. This merging process focuses on covering semantic gaps by exploring semantic connections among the patches, such as the influence area of each patch where the potential interaction with other patches may exist; and applying visual organization rules of proximity property and color harmony property. The proposed method is simple yet efficient computationally. It does not have any limitation on the input image class and does not include a training process. Our experiments indicate that it handles complex scenes with clutters and occlusions very well, and outperforms general segmentation methods by its meaningful semantic grouping and self determined stop criteria. © 2011 IEEE.
dc.description.nature	published_or_final_version
dc.description.other	The 2011 IEEE International Conference on Multimedia and Expo (ICME 2011), Barcelona, Spain, 11-15 July 2011. In Conference Proceedings of ICME, 2011, p. 1-6
dc.identifier.citation	The 2011 IEEE International Conference on Multimedia and Expo (ICME 2011), Barcelona, Spain, 11-15 July 2011. In Conference Proceedings of ICME, 2011, p. 1-6 [How to Cite?] DOI: http://dx.doi.org/10.1109/ICME.2011.6011910
dc.identifier.doi	http://dx.doi.org/10.1109/ICME.2011.6011910
dc.identifier.epage	6
dc.identifier.hkuros	201096
dc.identifier.isbn	978-1-61284-350-6
dc.identifier.scopus	eid_2-s2.0-80155132496
dc.identifier.spage	1
dc.identifier.uri	http://hdl.handle.net/10722/158735
dc.language	eng
dc.publisher	IEEE. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000477
dc.publisher.place	United States
dc.relation.ispartof	IEEE International Conference on Multimedia and Expo Proceedings
dc.relation.references	References in Scopus
dc.rights	IEEE International Conference on Multimedia and Expo. Copyright © IEEE.
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.rights	Creative Commons: Attribution 3.0 Hong Kong License
dc.subject	Sub-scene
dc.subject	Semantic gap
dc.subject	Complex scene
dc.subject	Color harmony
dc.subject	Image patches
dc.title	Sub-scene generation: a step towards complex scene understanding
dc.type	Conference_Paper

<?xml encoding="utf-8" version="1.0"?><item><contributor.author>Zhu, S</contributor.author><contributor.author>Yung, NHC</contributor.author><date.accessioned>2012-08-08T09:01:05Z</date.accessioned><date.available>2012-08-08T09:01:05Z</date.available><date.issued>2011</date.issued><identifier.citation>The 2011 IEEE International Conference on Multimedia and Expo (ICME 2011), Barcelona, Spain, 11-15 July 2011. In Conference Proceedings of ICME, 2011, p. 1-6</identifier.citation><identifier.isbn>978-1-61284-350-6</identifier.isbn><identifier.uri>http://hdl.handle.net/10722/158735</identifier.uri><description.abstract>This paper proposes a new method for generating sub-scenes from images of natural scenes. It is based on over-segmented image patches initially and uses rules derived from human psychology to merge the patches into reasonable sub-scenes. This merging process focuses on covering semantic gaps by exploring semantic connections among the patches, such as the influence area of each patch where the potential interaction with other patches may exist; and applying visual organization rules of proximity property and color harmony property. The proposed method is simple yet efficient computationally. It does not have any limitation on the input image class and does not include a training process. Our experiments indicate that it handles complex scenes with clutters and occlusions very well, and outperforms general segmentation methods by its meaningful semantic grouping and self determined stop criteria. &#169; 2011 IEEE.</description.abstract><language>eng</language><publisher>IEEE. The Journal&apos;s web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000477</publisher><relation.ispartof>IEEE International Conference on Multimedia and Expo Proceedings</relation.ispartof><rights>IEEE International Conference on Multimedia and Expo. Copyright &#169; IEEE.</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>Sub-scene</subject><subject>Semantic gap</subject><subject>Complex scene</subject><subject>Color harmony</subject><subject>Image patches</subject><title>Sub-scene generation: a step towards complex scene understanding</title><type>Conference_Paper</type><description.nature>published_or_final_version</description.nature><identifier.doi>10.1109/ICME.2011.6011910</identifier.doi><identifier.scopus>eid_2-s2.0-80155132496</identifier.scopus><identifier.hkuros>201096</identifier.hkuros><relation.references>http://www.scopus.com/mlt/select.url?eid=2-s2.0-80155132496&amp;selection=ref&amp;src=s&amp;origin=recordpage</relation.references><identifier.spage>1</identifier.spage><identifier.epage>6</identifier.epage><publisher.place>United States</publisher.place><description.other>The 2011 IEEE International Conference on Multimedia and Expo (ICME 2011), Barcelona, Spain, 11-15 July 2011. In Conference Proceedings of ICME, 2011, p. 1-6</description.other><bitstream.url>http://hub.hku.hk/bitstream/10722/158735/1/Content.pdf</bitstream.url></item>

Author Affiliations

The University of Hong Kong