Loading…

Transferable Contrastive Network for Generalized Zero-Shot Learning

Zero-shot learning (ZSL) is a challenging problem that aims to recognize the target categories without seen data, where semantic information is leveraged to transfer knowledge from some source classes. Although ZSL has made great progress in recent years, most existing approaches are easy to overfit...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiang, Huajie, Wang, Ruiping, Shan, Shiguang, Chen, Xilin
Format: Conference Proceeding
Language:English
Subjects:
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c315t-73513d27e9e3bac4630aecda22f21e3128987beece36521ce7f61a9f7a309f403
cites
container_end_page 9773
container_issue
container_start_page 9764
container_title
container_volume
creator Jiang, Huajie
Wang, Ruiping
Shan, Shiguang
Chen, Xilin
description Zero-shot learning (ZSL) is a challenging problem that aims to recognize the target categories without seen data, where semantic information is leveraged to transfer knowledge from some source classes. Although ZSL has made great progress in recent years, most existing approaches are easy to overfit the sources classes in generalized zero-shot learning (GZSL) task, which indicates that they learn little knowledge about target classes. To tackle such problem, we propose a novel Transferable Contrastive Network (TCN) that explicitly transfers knowledge from the source classes to the target classes. It automatically contrasts one image with different classes to judge whether they are consistent or not. By exploiting the class similarities to make knowledge transfer from source images to similar target classes, our approach is more robust to recognize the target images. Experiments on five benchmark datasets show the superiority of our approach for GZSL.
doi_str_mv 10.1109/ICCV.2019.00986
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_9009572</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9009572</ieee_id><sourcerecordid>9009572</sourcerecordid><originalsourceid>FETCH-LOGICAL-c315t-73513d27e9e3bac4630aecda22f21e3128987beece36521ce7f61a9f7a309f403</originalsourceid><addsrcrecordid>eNotj8FKxDAUAKMguK579uAlP9D6Xl7bJEcJui4UPbh68LKk7YtWayppUfTrXdDTXIaBEeIMIUcEe7Fx7jFXgDYHsKY6ECurDWplsDBA5lAsFBnIdAnFsTiZplcAsspUC-G2yccpcPLNwNKNcU5-mvtPlrc8f43pTYYxyTXHvTH0P9zJJ05jdv8yzrJmn2Ifn0_FUfDDxKt_LsXD9dXW3WT13XrjLuusJSznTFOJ1CnNlqnxbVEReG47r1RQyITKWKMb5papKhW2rEOF3gbtCWwogJbi_K_bM_PuI_XvPn3v7H651Ip-AbTeSpU</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Transferable Contrastive Network for Generalized Zero-Shot Learning</title><source>IEEE Xplore All Conference Series</source><creator>Jiang, Huajie ; Wang, Ruiping ; Shan, Shiguang ; Chen, Xilin</creator><creatorcontrib>Jiang, Huajie ; Wang, Ruiping ; Shan, Shiguang ; Chen, Xilin</creatorcontrib><description>Zero-shot learning (ZSL) is a challenging problem that aims to recognize the target categories without seen data, where semantic information is leveraged to transfer knowledge from some source classes. Although ZSL has made great progress in recent years, most existing approaches are easy to overfit the sources classes in generalized zero-shot learning (GZSL) task, which indicates that they learn little knowledge about target classes. To tackle such problem, we propose a novel Transferable Contrastive Network (TCN) that explicitly transfers knowledge from the source classes to the target classes. It automatically contrasts one image with different classes to judge whether they are consistent or not. By exploiting the class similarities to make knowledge transfer from source images to similar target classes, our approach is more robust to recognize the target images. Experiments on five benchmark datasets show the superiority of our approach for GZSL.</description><identifier>EISSN: 2380-7504</identifier><identifier>EISBN: 9781728148038</identifier><identifier>EISBN: 1728148030</identifier><identifier>DOI: 10.1109/ICCV.2019.00986</identifier><language>eng</language><publisher>IEEE</publisher><subject>Fuses ; Image recognition ; Robustness ; Semantics ; Target recognition ; Task analysis ; Visualization</subject><ispartof>2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, p.9764-9773</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c315t-73513d27e9e3bac4630aecda22f21e3128987beece36521ce7f61a9f7a309f403</citedby></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9009572$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,27925,54555,54932</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9009572$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Jiang, Huajie</creatorcontrib><creatorcontrib>Wang, Ruiping</creatorcontrib><creatorcontrib>Shan, Shiguang</creatorcontrib><creatorcontrib>Chen, Xilin</creatorcontrib><title>Transferable Contrastive Network for Generalized Zero-Shot Learning</title><title>2019 IEEE/CVF International Conference on Computer Vision (ICCV)</title><addtitle>ICCV</addtitle><description>Zero-shot learning (ZSL) is a challenging problem that aims to recognize the target categories without seen data, where semantic information is leveraged to transfer knowledge from some source classes. Although ZSL has made great progress in recent years, most existing approaches are easy to overfit the sources classes in generalized zero-shot learning (GZSL) task, which indicates that they learn little knowledge about target classes. To tackle such problem, we propose a novel Transferable Contrastive Network (TCN) that explicitly transfers knowledge from the source classes to the target classes. It automatically contrasts one image with different classes to judge whether they are consistent or not. By exploiting the class similarities to make knowledge transfer from source images to similar target classes, our approach is more robust to recognize the target images. Experiments on five benchmark datasets show the superiority of our approach for GZSL.</description><subject>Fuses</subject><subject>Image recognition</subject><subject>Robustness</subject><subject>Semantics</subject><subject>Target recognition</subject><subject>Task analysis</subject><subject>Visualization</subject><issn>2380-7504</issn><isbn>9781728148038</isbn><isbn>1728148030</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2019</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotj8FKxDAUAKMguK579uAlP9D6Xl7bJEcJui4UPbh68LKk7YtWayppUfTrXdDTXIaBEeIMIUcEe7Fx7jFXgDYHsKY6ECurDWplsDBA5lAsFBnIdAnFsTiZplcAsspUC-G2yccpcPLNwNKNcU5-mvtPlrc8f43pTYYxyTXHvTH0P9zJJ05jdv8yzrJmn2Ifn0_FUfDDxKt_LsXD9dXW3WT13XrjLuusJSznTFOJ1CnNlqnxbVEReG47r1RQyITKWKMb5papKhW2rEOF3gbtCWwogJbi_K_bM_PuI_XvPn3v7H651Ip-AbTeSpU</recordid><startdate>20191001</startdate><enddate>20191001</enddate><creator>Jiang, Huajie</creator><creator>Wang, Ruiping</creator><creator>Shan, Shiguang</creator><creator>Chen, Xilin</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>20191001</creationdate><title>Transferable Contrastive Network for Generalized Zero-Shot Learning</title><author>Jiang, Huajie ; Wang, Ruiping ; Shan, Shiguang ; Chen, Xilin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c315t-73513d27e9e3bac4630aecda22f21e3128987beece36521ce7f61a9f7a309f403</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Fuses</topic><topic>Image recognition</topic><topic>Robustness</topic><topic>Semantics</topic><topic>Target recognition</topic><topic>Task analysis</topic><topic>Visualization</topic><toplevel>online_resources</toplevel><creatorcontrib>Jiang, Huajie</creatorcontrib><creatorcontrib>Wang, Ruiping</creatorcontrib><creatorcontrib>Shan, Shiguang</creatorcontrib><creatorcontrib>Chen, Xilin</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Jiang, Huajie</au><au>Wang, Ruiping</au><au>Shan, Shiguang</au><au>Chen, Xilin</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Transferable Contrastive Network for Generalized Zero-Shot Learning</atitle><btitle>2019 IEEE/CVF International Conference on Computer Vision (ICCV)</btitle><stitle>ICCV</stitle><date>2019-10-01</date><risdate>2019</risdate><spage>9764</spage><epage>9773</epage><pages>9764-9773</pages><eissn>2380-7504</eissn><eisbn>9781728148038</eisbn><eisbn>1728148030</eisbn><abstract>Zero-shot learning (ZSL) is a challenging problem that aims to recognize the target categories without seen data, where semantic information is leveraged to transfer knowledge from some source classes. Although ZSL has made great progress in recent years, most existing approaches are easy to overfit the sources classes in generalized zero-shot learning (GZSL) task, which indicates that they learn little knowledge about target classes. To tackle such problem, we propose a novel Transferable Contrastive Network (TCN) that explicitly transfers knowledge from the source classes to the target classes. It automatically contrasts one image with different classes to judge whether they are consistent or not. By exploiting the class similarities to make knowledge transfer from source images to similar target classes, our approach is more robust to recognize the target images. Experiments on five benchmark datasets show the superiority of our approach for GZSL.</abstract><pub>IEEE</pub><doi>10.1109/ICCV.2019.00986</doi><tpages>10</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2380-7504
ispartof 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, p.9764-9773
issn 2380-7504
language eng
recordid cdi_ieee_primary_9009572
source IEEE Xplore All Conference Series
subjects Fuses
Image recognition
Robustness
Semantics
Target recognition
Task analysis
Visualization
title Transferable Contrastive Network for Generalized Zero-Shot Learning
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T22%3A03%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Transferable%20Contrastive%20Network%20for%20Generalized%20Zero-Shot%20Learning&rft.btitle=2019%20IEEE/CVF%20International%20Conference%20on%20Computer%20Vision%20(ICCV)&rft.au=Jiang,%20Huajie&rft.date=2019-10-01&rft.spage=9764&rft.epage=9773&rft.pages=9764-9773&rft.eissn=2380-7504&rft_id=info:doi/10.1109/ICCV.2019.00986&rft.eisbn=9781728148038&rft.eisbn_list=1728148030&rft_dat=%3Cieee_CHZPO%3E9009572%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c315t-73513d27e9e3bac4630aecda22f21e3128987beece36521ce7f61a9f7a309f403%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=9009572&rfr_iscdi=true