Loading…
Learning Representative Deep Features for Image Set Analysis
This paper proposes to learn features from sets of labeled raw images. With this method, the problem of over-fitting can be effectively suppressed, so that deep CNNs can be trained from scratch with a small number of training data, i.e., 420 labeled albums with about 30 000 photos. This method can e...
Saved in:
Published in: | IEEE transactions on multimedia 2015-11, Vol.17 (11), p.1960-1968 |
---|---|
Main Authors: | , , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites Items that cite this one |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | cdi_FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003 |
---|---|
cites | cdi_FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003 |
container_end_page | 1968 |
container_issue | 11 |
container_start_page | 1960 |
container_title | IEEE transactions on multimedia |
container_volume | 17 |
creator | Wu, Zifeng Huang, Yongzhen Wang, Liang |
description | This paper proposes to learn features from sets of labeled raw images. With this method, the problem of over-fitting can be effectively suppressed, so that deep CNNs can be trained from scratch with a small number of training data, i.e., 420 labeled albums with about 30 000 photos. This method can effectively deal with sets of images, no matter if the sets bear temporal structures. A typical approach to sequential image analysis usually leverages motions between adjacent frames, while the proposed method focuses on capturing the co-occurrences and frequencies of features. Nevertheless, our method outperforms previous best performers in terms of album classification, and achieves comparable or even better performances in terms of gait based human identification. These results demonstrate its effectiveness and good adaptivity to different kinds of set data. |
doi_str_mv | 10.1109/TMM.2015.2477681 |
format | article |
fullrecord | <record><control><sourceid>proquest_ieee_</sourceid><recordid>TN_cdi_proquest_miscellaneous_1778007760</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>7254176</ieee_id><sourcerecordid>3856168831</sourcerecordid><originalsourceid>FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003</originalsourceid><addsrcrecordid>eNpdkNFLwzAQh4MoOKfvgi8FX3zpzCVpsoIvYzoddAg6n0OaXUZH19akFfbf27Lhg093HN_vuPsIuQU6AaDp43q1mjAKyYQJpeQUzsgIUgExpUqd933CaJwyoJfkKoQdpSASqkbkKUPjq6LaRh_YeAxYtaYtfjB6RmyiBZq266eRq3203JstRp_YRrPKlIdQhGty4UwZ8OZUx-Rr8bKev8XZ--tyPstiyxW08UaJPOeI1pkkdQmXSgIytrHCWSYYWgSZ5k5RZCC4EFY6wXPDzQaYE5TyMXk47m18_d1haPW-CBbL0lRYd0GDUtPhTzmg9__QXd35_t6BYilPhRSip-iRsr4OwaPTjS_2xh80UD3o1L1OPejUJ5195O4YKRDxD1csEaAk_wX2jG9i</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1729394644</pqid></control><display><type>article</type><title>Learning Representative Deep Features for Image Set Analysis</title><source>IEEE Electronic Library (IEL) Journals</source><creator>Wu, Zifeng ; Huang, Yongzhen ; Wang, Liang</creator><creatorcontrib>Wu, Zifeng ; Huang, Yongzhen ; Wang, Liang</creatorcontrib><description>This paper proposes to learn features from sets of labeled raw images. With this method, the problem of over-fitting can be effectively suppressed, so that deep CNNs can be trained from scratch with a small number of training data, i.e., 420 labeled albums with about 30 000 photos. This method can effectively deal with sets of images, no matter if the sets bear temporal structures. A typical approach to sequential image analysis usually leverages motions between adjacent frames, while the proposed method focuses on capturing the co-occurrences and frequencies of features. Nevertheless, our method outperforms previous best performers in terms of album classification, and achieves comparable or even better performances in terms of gait based human identification. These results demonstrate its effectiveness and good adaptivity to different kinds of set data.</description><identifier>ISSN: 1520-9210</identifier><identifier>EISSN: 1941-0077</identifier><identifier>DOI: 10.1109/TMM.2015.2477681</identifier><identifier>CODEN: ITMUF8</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Album classification ; Classification ; Convolution ; Data models ; deep learning ; Feature extraction ; Fittings ; gait recognition ; Hidden Markov models ; Human performance ; image set ; Learning ; Multimedia ; Temporal logic ; Training ; Training data ; Videos</subject><ispartof>IEEE transactions on multimedia, 2015-11, Vol.17 (11), p.1960-1968</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Nov 2015</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003</citedby><cites>FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/7254176$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,54796</link.rule.ids></links><search><creatorcontrib>Wu, Zifeng</creatorcontrib><creatorcontrib>Huang, Yongzhen</creatorcontrib><creatorcontrib>Wang, Liang</creatorcontrib><title>Learning Representative Deep Features for Image Set Analysis</title><title>IEEE transactions on multimedia</title><addtitle>TMM</addtitle><description>This paper proposes to learn features from sets of labeled raw images. With this method, the problem of over-fitting can be effectively suppressed, so that deep CNNs can be trained from scratch with a small number of training data, i.e., 420 labeled albums with about 30 000 photos. This method can effectively deal with sets of images, no matter if the sets bear temporal structures. A typical approach to sequential image analysis usually leverages motions between adjacent frames, while the proposed method focuses on capturing the co-occurrences and frequencies of features. Nevertheless, our method outperforms previous best performers in terms of album classification, and achieves comparable or even better performances in terms of gait based human identification. These results demonstrate its effectiveness and good adaptivity to different kinds of set data.</description><subject>Album classification</subject><subject>Classification</subject><subject>Convolution</subject><subject>Data models</subject><subject>deep learning</subject><subject>Feature extraction</subject><subject>Fittings</subject><subject>gait recognition</subject><subject>Hidden Markov models</subject><subject>Human performance</subject><subject>image set</subject><subject>Learning</subject><subject>Multimedia</subject><subject>Temporal logic</subject><subject>Training</subject><subject>Training data</subject><subject>Videos</subject><issn>1520-9210</issn><issn>1941-0077</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2015</creationdate><recordtype>article</recordtype><recordid>eNpdkNFLwzAQh4MoOKfvgi8FX3zpzCVpsoIvYzoddAg6n0OaXUZH19akFfbf27Lhg093HN_vuPsIuQU6AaDp43q1mjAKyYQJpeQUzsgIUgExpUqd933CaJwyoJfkKoQdpSASqkbkKUPjq6LaRh_YeAxYtaYtfjB6RmyiBZq266eRq3203JstRp_YRrPKlIdQhGty4UwZ8OZUx-Rr8bKev8XZ--tyPstiyxW08UaJPOeI1pkkdQmXSgIytrHCWSYYWgSZ5k5RZCC4EFY6wXPDzQaYE5TyMXk47m18_d1haPW-CBbL0lRYd0GDUtPhTzmg9__QXd35_t6BYilPhRSip-iRsr4OwaPTjS_2xh80UD3o1L1OPejUJ5195O4YKRDxD1csEaAk_wX2jG9i</recordid><startdate>201511</startdate><enddate>201511</enddate><creator>Wu, Zifeng</creator><creator>Huang, Yongzhen</creator><creator>Wang, Liang</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>F28</scope><scope>FR3</scope></search><sort><creationdate>201511</creationdate><title>Learning Representative Deep Features for Image Set Analysis</title><author>Wu, Zifeng ; Huang, Yongzhen ; Wang, Liang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2015</creationdate><topic>Album classification</topic><topic>Classification</topic><topic>Convolution</topic><topic>Data models</topic><topic>deep learning</topic><topic>Feature extraction</topic><topic>Fittings</topic><topic>gait recognition</topic><topic>Hidden Markov models</topic><topic>Human performance</topic><topic>image set</topic><topic>Learning</topic><topic>Multimedia</topic><topic>Temporal logic</topic><topic>Training</topic><topic>Training data</topic><topic>Videos</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Wu, Zifeng</creatorcontrib><creatorcontrib>Huang, Yongzhen</creatorcontrib><creatorcontrib>Wang, Liang</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Xplore</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>ANTE: Abstracts in New Technology & Engineering</collection><collection>Engineering Research Database</collection><jtitle>IEEE transactions on multimedia</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Wu, Zifeng</au><au>Huang, Yongzhen</au><au>Wang, Liang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Learning Representative Deep Features for Image Set Analysis</atitle><jtitle>IEEE transactions on multimedia</jtitle><stitle>TMM</stitle><date>2015-11</date><risdate>2015</risdate><volume>17</volume><issue>11</issue><spage>1960</spage><epage>1968</epage><pages>1960-1968</pages><issn>1520-9210</issn><eissn>1941-0077</eissn><coden>ITMUF8</coden><abstract>This paper proposes to learn features from sets of labeled raw images. With this method, the problem of over-fitting can be effectively suppressed, so that deep CNNs can be trained from scratch with a small number of training data, i.e., 420 labeled albums with about 30 000 photos. This method can effectively deal with sets of images, no matter if the sets bear temporal structures. A typical approach to sequential image analysis usually leverages motions between adjacent frames, while the proposed method focuses on capturing the co-occurrences and frequencies of features. Nevertheless, our method outperforms previous best performers in terms of album classification, and achieves comparable or even better performances in terms of gait based human identification. These results demonstrate its effectiveness and good adaptivity to different kinds of set data.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/TMM.2015.2477681</doi><tpages>9</tpages></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1520-9210 |
ispartof | IEEE transactions on multimedia, 2015-11, Vol.17 (11), p.1960-1968 |
issn | 1520-9210 1941-0077 |
language | eng |
recordid | cdi_proquest_miscellaneous_1778007760 |
source | IEEE Electronic Library (IEL) Journals |
subjects | Album classification Classification Convolution Data models deep learning Feature extraction Fittings gait recognition Hidden Markov models Human performance image set Learning Multimedia Temporal logic Training Training data Videos |
title | Learning Representative Deep Features for Image Set Analysis |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-06T20%3A34%3A00IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_ieee_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Learning%20Representative%20Deep%20Features%20for%20Image%20Set%20Analysis&rft.jtitle=IEEE%20transactions%20on%20multimedia&rft.au=Wu,%20Zifeng&rft.date=2015-11&rft.volume=17&rft.issue=11&rft.spage=1960&rft.epage=1968&rft.pages=1960-1968&rft.issn=1520-9210&rft.eissn=1941-0077&rft.coden=ITMUF8&rft_id=info:doi/10.1109/TMM.2015.2477681&rft_dat=%3Cproquest_ieee_%3E3856168831%3C/proquest_ieee_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c371t-d74bb3eecfa59f536761e22dc4fc242ece169bf70e214344c6f43ba3ad12f4003%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1729394644&rft_id=info:pmid/&rft_ieee_id=7254176&rfr_iscdi=true |