On the Relationship between Visual Attributes and Convolutional Networks

Handle URI:
http://hdl.handle.net/10754/556138
Title:
On the Relationship between Visual Attributes and Convolutional Networks
Authors:
Castillo, Victor; Ghanem, Bernard ( 0000-0002-5534-587X ) ; Niebles, Juan Carlos
Abstract:
One of the cornerstone principles of deep models is their abstraction capacity, i.e. their ability to learn abstract concepts from ‘simpler’ ones. Through extensive experiments, we characterize the nature of the relationship between abstract concepts (specifically objects in images) learned by popular and high performing convolutional networks (conv-nets) and established mid-level representations used in computer vision (specifically semantic visual attributes). We focus on attributes due to their impact on several applications, such as object description, retrieval and mining, and active (and zero-shot) learning. Among the findings we uncover, we show empirical evidence of the existence of Attribute Centric Nodes (ACNs) within a conv-net, which is trained to recognize objects (not attributes) in images. These special conv-net nodes (1) collectively encode information pertinent to visual attribute representation and discrimination, (2) are unevenly and sparsely distribution across all layers of the conv-net, and (3) play an important role in conv-net based object recognition.
KAUST Department:
Image and Video Understanding Lab
Publisher:
IEEE
Journal:
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition
Conference/Event name:
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Issue Date:
2-Jun-2015
Type:
Conference Paper
Sponsors:
IEEE Computer Society, Computer Vision Foundation - CVF
Additional Links:
https://dl.dropboxusercontent.com/u/18955644/website_files/attributes_and_CNNs_CVPR2015.pdf; https://dl.dropboxusercontent.com/u/18955644/website_files/0052-supp.zip
Appears in Collections:
Conference Papers

Full metadata record

DC FieldValue Language
dc.contributor.authorCastillo, Victoren
dc.contributor.authorGhanem, Bernarden
dc.contributor.authorNiebles, Juan Carlosen
dc.date.accessioned2015-06-02T13:29:58Zen
dc.date.available2015-06-02T13:29:58Zen
dc.date.issued2015-06-02en
dc.identifier.urihttp://hdl.handle.net/10754/556138en
dc.description.abstractOne of the cornerstone principles of deep models is their abstraction capacity, i.e. their ability to learn abstract concepts from ‘simpler’ ones. Through extensive experiments, we characterize the nature of the relationship between abstract concepts (specifically objects in images) learned by popular and high performing convolutional networks (conv-nets) and established mid-level representations used in computer vision (specifically semantic visual attributes). We focus on attributes due to their impact on several applications, such as object description, retrieval and mining, and active (and zero-shot) learning. Among the findings we uncover, we show empirical evidence of the existence of Attribute Centric Nodes (ACNs) within a conv-net, which is trained to recognize objects (not attributes) in images. These special conv-net nodes (1) collectively encode information pertinent to visual attribute representation and discrimination, (2) are unevenly and sparsely distribution across all layers of the conv-net, and (3) play an important role in conv-net based object recognition.en
dc.description.sponsorshipIEEE Computer Society, Computer Vision Foundation - CVFen
dc.publisherIEEEen
dc.relation.urlhttps://dl.dropboxusercontent.com/u/18955644/website_files/attributes_and_CNNs_CVPR2015.pdfen
dc.relation.urlhttps://dl.dropboxusercontent.com/u/18955644/website_files/0052-supp.zipen
dc.rights(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.en
dc.subjectVisual Attributesen
dc.subjectConvolution Networksen
dc.titleOn the Relationship between Visual Attributes and Convolutional Networksen
dc.typeConference Paperen
dc.contributor.departmentImage and Video Understanding Laben
dc.identifier.journalProceedings of the IEEE Conference on Computer Vision and Pattern Recognitionen
dc.conference.date07 Jun - 12 Jun 2015en
dc.conference.name2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)en
dc.conference.locationHynes Convention Center 900 Boylston St Boston, MA, USAen
dc.eprint.versionPost-printen
dc.contributor.institutionUniversidad del Norte, Colombiaen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.