Staff View
Toward a Taxonomy and Computational Models of Abnormalities in Images

Descriptive

TypeOfResource
Text
TitleInfo
Title
Toward a Taxonomy and Computational Models of Abnormalities in Images
Name (type = personal)
NamePart (type = family)
Saleh
NamePart (type = given)
Babak
Affiliation
Computer Science (New Brunswick), Rutgers University
Role
RoleTerm (authority = marcrt); (type = text)
author
Name (type = personal)
NamePart (type = family)
Elgammal
NamePart (type = given)
Ahmed
Affiliation
Computer Science (New Brunswick), Rutgers University
Role
RoleTerm (authority = marcrt); (type = text)
author
Name (type = personal)
NamePart (type = family)
Feldman
NamePart (type = given)
Jacob
Affiliation
Psychology (New Brunswick), Rutgers University
Role
RoleTerm (authority = marcrt); (type = text)
author
Name (type = personal)
NamePart (type = family)
Farhadi
NamePart (type = given)
Ali
Affiliation
University of Washington
Role
RoleTerm (authority = marcrt); (type = text)
author
Name (authority = RutgersOrg-School); (type = corporate)
NamePart
School of Arts and Sciences (SAS) (New Brunswick)
Name (authority = RutgersOrg-Department); (type = corporate)
NamePart
Computer Science (New Brunswick)
Name (authority = RutgersOrg-Department); (type = corporate)
NamePart
Psychology (New Brunswick)
Genre (authority = RULIB-FS)
Conference Paper or Lecture
Genre (authority = NISO JAV)
Accepted Manuscript (AM)
OriginInfo
DateCreated (encoding = w3cdtf); (keyDate = yes); (qualifier = exact)
2015
Language
LanguageTerm (authority = ISO 639-3:2007); (type = text)
English
PhysicalDescription
InternetMediaType
application/pdf
Extent
9 p.
Abstract (type = abstract)
The human visual system can spot an abnormal image, and reason about what makes it strange. This task has not received enough attention in computer vision. In this paper we study various types of atypicalities in images in a more comprehensive way than has been done before. We propose a new dataset of abnormal images showing a wide range of atypicalities. We design human subject experiments to discover a coarse taxonomy of the reasons for abnormality. Our experiments reveal three major categories of abnormality: object-centric, scene-centric, and contextual. Based on this taxonomy, we propose a comprehensive computational model that can predict all different types of abnormality in images and outperform prior arts in abnormality recognition.
Extension
DescriptiveEvent
Type
Citation
DateTime (encoding = w3cdtf)
2016
AssociatedObject
Type
Journal
Relationship
Has part
Name
Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence and the Twenty-Eighth Innovative Applications of Artificial Intelligence Conference
Reference (type = url)
http://www.aaai.org/Press/Proceedings/aaai16.php
AssociatedEntity
Role
Publisher
Name
Association for the Advancement of Artificial Intelligence
Extension
DescriptiveEvent
Type
Conference
Label
AAAI-16: 30th AAAI Conference on Artificial Intelligence
Place
Phoenix, AZ
DateTime (encoding = w3cdtf)
2016-02
AssociatedEntity
Role
Sponsor
Name
Association for the Advancement of Artificial Intelligence
Note (type = peerReview)
Peer reviewed
Subject (authority = LCSH)
Topic
Computer vision
Subject (authority = local)
Topic
Abnormal images
Subject (authority = LCSH)
Topic
Image processing
Subject (authority = local)
Topic
Taxonomies
RelatedItem (type = host)
TitleInfo
Title
Elgammal, Ahmed
Identifier (type = local)
rucore30188000001
RelatedItem (type = host)
TitleInfo
Title
Feldman, Jacob
Identifier (type = local)
rucore30178100001
Extension
DescriptiveEvent
Type
Grant award
AssociatedEntity
Role
Funder
Name
National Science Foundation
AssociatedObject
Type
Grant number
Name
IIS-1218872
RelatedItem (type = host)
TitleInfo
Title
Saleh, Babak
Identifier (type = local)
rucore30180600001
Location
PhysicalLocation (authority = marcorg); (displayLabel = Rutgers, The State University of New Jersey)
NjNbRU
Identifier (type = doi)
doi:10.7282/T3319XTQ
Back to the top

Rights

RightsDeclaration (AUTHORITY = FS); (ID = rulibRdec0004)
Copyright for scholarly resources published in RUcore is retained by the copyright holder. By virtue of its appearance in this open access medium, you are free to use this resource, with proper attribution, in educational and other non-commercial settings. Other uses, such as reproduction or republication, may require the permission of the copyright holder.
Copyright
Status
Copyright protected
Availability
Status
Open
Reason
Permission or license
RightsEvent
Type
Permission or license
AssociatedObject
Type
License
Name
Multiple author license v. 1
Detail
I hereby grant to Rutgers, The State University of New Jersey (Rutgers) the non-exclusive right to retain, reproduce, and distribute the deposited work (Work) in whole or in part, in and from its electronic format, without fee. This agreement does not represent a transfer of copyright to Rutgers.Rutgers may make and keep more than one copy of the Work for purposes of security, backup, preservation, and access and may migrate the Work to any medium or format for the purpose of preservation and access in the future. Rutgers will not make any alteration, other than as allowed by this agreement, to the Work.I represent and warrant to Rutgers that the Work is my original work. I also represent that the Work does not, to the best of my knowledge, infringe or violate any rights of others.I further represent and warrant that I have obtained all necessary rights to permit Rutgers to reproduce and distribute the Work and that any third-party owned content is clearly identified and acknowledged within the Work.By granting this license, I acknowledge that I have read and agreed to the terms of this agreement and all related RUcore and Rutgers policies.
RightsEvent
Type
Embargo
DateTime (point = start); (encoding = w3cdtf)
2015-12-02
DateTime (point = end); (encoding = w3cdtf)
2016-02-28
Detail
Access to this PDF is restricted in accordance with the policies of AAAI, which permits only limited distribution prior to AAAI publication. Access to this version will be made available following the AAAI Conference in February 2016.
RightsEvent
Type
Permissions research
DateTime (encoding = w3cdtf)
2015-12-03
AssociatedEntity
Role
Cataloger
Name
Rhonda Marker
Detail
Author's toolkit here: http://www.aaai.org/Conferences/AAAI/2016/aaai16call.php. Copyright form allows author to post on employer's own web page or ftp site, but only limited distribution of the article/paper prior to publication. Distribution license grants AAAI nonexclusive rights to use the work.
RightsHolder (type = corporate)
Name
Association for the Advancement of Artificial Intelligence
Role
Copyright holder
Back to the top

Technical

RULTechMD (ID = TECHNICAL1)
ContentModel
Document
Back to the top
Version 8.3.10
Rutgers University Libraries - Copyright ©2019