Variational Multi-Prototype Encoder for Object Recognition Using Multiple Prototype Images
In the recent research of Variational Prototyping-Encoder (VPE), the problem of classifying 2D flat objects of the unseen class has been addressed. VPE solves this problem by pre-learning the image translation task from real-world object images to their corresponding prototype images as a meta-task....
Gespeichert in:
Veröffentlicht in: | IEEE access 2022, Vol.10, p.19586-19598 |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 19598 |
---|---|
container_issue | |
container_start_page | 19586 |
container_title | IEEE access |
container_volume | 10 |
creator | Kang, Jun Seok Ahn, Sang Chul |
description | In the recent research of Variational Prototyping-Encoder (VPE), the problem of classifying 2D flat objects of the unseen class has been addressed. VPE solves this problem by pre-learning the image translation task from real-world object images to their corresponding prototype images as a meta-task. VPE uses a single prototype for each object class. However, in general, a single prototype is not sufficient to represent a generic object class because the appearance can change significantly according to viewpoints and other factors. In this case, using VPE and a single prototype for each class in training can result in overfitting or performance degradation. One solution may be the use of multiple prototypes. However, this also requires costly sub-labeling for dividing the input class into smaller classes and assigning a prototype to each. Therefore, we propose a new learning method, the variational multi-prototype encoder (VaMPE), which can overcome the limitations of VPE and use multiple prototypes for each object class. The proposed method does not require additional sub-labeling other than simply adding multiple prototypes to each class. Through various experiments, we demonstrate that the proposed method outperforms VPE. |
doi_str_mv | 10.1109/ACCESS.2022.3151856 |
format | Article |
fullrecord | <record><control><sourceid>proquest_doaj_</sourceid><recordid>TN_cdi_proquest_journals_2633041944</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9714291</ieee_id><doaj_id>oai_doaj_org_article_cdc0924d57624fa1b5e1a0813fd95b4b</doaj_id><sourcerecordid>2633041944</sourcerecordid><originalsourceid>FETCH-LOGICAL-c358t-46fba21773555e917d629e8cfb46db09872929a7d98c8d9c9d00205e2956c4aa3</originalsourceid><addsrcrecordid>eNpNkU9Lw0AQxYMoWGo_QS8Bz6n7P9ljKVULlYq1Hrwsm91J2ZJ242566Lc3NVKdywyP-b2BeUkyxmiCMZIP09lsvl5PCCJkQjHHBRdXyYBgITPKqbj-N98moxh3qKuik3g-SD4_dHC6df6g6_TlWLcuew2-9e2pgXR-MN5CSCsf0lW5A9Omb2D89uDOQLqJ7rDtoaaG9I9b7PUW4l1yU-k6wui3D5PN4_x99pwtV0-L2XSZGcqLNmOiKjXBeU455yBxbgWRUJiqZMKWSBY5kUTq3MrCFFYaaREiiAORXBimNR0mi97Xer1TTXB7HU7Ka6d-BB-2SofWmRqUsQZJwizPBWGVxiUHrLtf0MpKXrKy87rvvZrgv44QW7Xzx9D9JioiKEUMS8a6LdpvmeBjDFBdrmKkzpmoPhN1zkT9ZtJR455yAHAhZI4ZkZh-A9zih70</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2633041944</pqid></control><display><type>article</type><title>Variational Multi-Prototype Encoder for Object Recognition Using Multiple Prototype Images</title><source>Directory of Open Access Journals</source><source>Free E-Journal (出版社公開部分のみ)</source><source>IEEE Xplore Open Access Journals</source><creator>Kang, Jun Seok ; Ahn, Sang Chul</creator><creatorcontrib>Kang, Jun Seok ; Ahn, Sang Chul</creatorcontrib><description>In the recent research of Variational Prototyping-Encoder (VPE), the problem of classifying 2D flat objects of the unseen class has been addressed. VPE solves this problem by pre-learning the image translation task from real-world object images to their corresponding prototype images as a meta-task. VPE uses a single prototype for each object class. However, in general, a single prototype is not sufficient to represent a generic object class because the appearance can change significantly according to viewpoints and other factors. In this case, using VPE and a single prototype for each class in training can result in overfitting or performance degradation. One solution may be the use of multiple prototypes. However, this also requires costly sub-labeling for dividing the input class into smaller classes and assigning a prototype to each. Therefore, we propose a new learning method, the variational multi-prototype encoder (VaMPE), which can overcome the limitations of VPE and use multiple prototypes for each object class. The proposed method does not require additional sub-labeling other than simply adding multiple prototypes to each class. Through various experiments, we demonstrate that the proposed method outperforms VPE.</description><identifier>ISSN: 2169-3536</identifier><identifier>EISSN: 2169-3536</identifier><identifier>DOI: 10.1109/ACCESS.2022.3151856</identifier><identifier>CODEN: IAECCG</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Coders ; Deep learning ; embedding space ; Feature extraction ; image classification ; Labeling ; Learning ; Neural networks ; Object recognition ; Performance degradation ; Perturbation methods ; prototype learning ; Prototypes ; Prototyping ; Task analysis ; Training ; variational encoder</subject><ispartof>IEEE access, 2022, Vol.10, p.19586-19598</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c358t-46fba21773555e917d629e8cfb46db09872929a7d98c8d9c9d00205e2956c4aa3</cites><orcidid>0000-0002-6384-891X ; 0000-0003-2959-1089</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9714291$$EHTML$$P50$$Gieee$$Hfree_for_read</linktohtml><link.rule.ids>314,780,784,864,2102,4024,27633,27923,27924,27925,54933</link.rule.ids></links><search><creatorcontrib>Kang, Jun Seok</creatorcontrib><creatorcontrib>Ahn, Sang Chul</creatorcontrib><title>Variational Multi-Prototype Encoder for Object Recognition Using Multiple Prototype Images</title><title>IEEE access</title><addtitle>Access</addtitle><description>In the recent research of Variational Prototyping-Encoder (VPE), the problem of classifying 2D flat objects of the unseen class has been addressed. VPE solves this problem by pre-learning the image translation task from real-world object images to their corresponding prototype images as a meta-task. VPE uses a single prototype for each object class. However, in general, a single prototype is not sufficient to represent a generic object class because the appearance can change significantly according to viewpoints and other factors. In this case, using VPE and a single prototype for each class in training can result in overfitting or performance degradation. One solution may be the use of multiple prototypes. However, this also requires costly sub-labeling for dividing the input class into smaller classes and assigning a prototype to each. Therefore, we propose a new learning method, the variational multi-prototype encoder (VaMPE), which can overcome the limitations of VPE and use multiple prototypes for each object class. The proposed method does not require additional sub-labeling other than simply adding multiple prototypes to each class. Through various experiments, we demonstrate that the proposed method outperforms VPE.</description><subject>Coders</subject><subject>Deep learning</subject><subject>embedding space</subject><subject>Feature extraction</subject><subject>image classification</subject><subject>Labeling</subject><subject>Learning</subject><subject>Neural networks</subject><subject>Object recognition</subject><subject>Performance degradation</subject><subject>Perturbation methods</subject><subject>prototype learning</subject><subject>Prototypes</subject><subject>Prototyping</subject><subject>Task analysis</subject><subject>Training</subject><subject>variational encoder</subject><issn>2169-3536</issn><issn>2169-3536</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>ESBDL</sourceid><sourceid>RIE</sourceid><sourceid>DOA</sourceid><recordid>eNpNkU9Lw0AQxYMoWGo_QS8Bz6n7P9ljKVULlYq1Hrwsm91J2ZJ242566Lc3NVKdywyP-b2BeUkyxmiCMZIP09lsvl5PCCJkQjHHBRdXyYBgITPKqbj-N98moxh3qKuik3g-SD4_dHC6df6g6_TlWLcuew2-9e2pgXR-MN5CSCsf0lW5A9Omb2D89uDOQLqJ7rDtoaaG9I9b7PUW4l1yU-k6wui3D5PN4_x99pwtV0-L2XSZGcqLNmOiKjXBeU455yBxbgWRUJiqZMKWSBY5kUTq3MrCFFYaaREiiAORXBimNR0mi97Xer1TTXB7HU7Ka6d-BB-2SofWmRqUsQZJwizPBWGVxiUHrLtf0MpKXrKy87rvvZrgv44QW7Xzx9D9JioiKEUMS8a6LdpvmeBjDFBdrmKkzpmoPhN1zkT9ZtJR455yAHAhZI4ZkZh-A9zih70</recordid><startdate>2022</startdate><enddate>2022</enddate><creator>Kang, Jun Seok</creator><creator>Ahn, Sang Chul</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>ESBDL</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7SR</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0002-6384-891X</orcidid><orcidid>https://orcid.org/0000-0003-2959-1089</orcidid></search><sort><creationdate>2022</creationdate><title>Variational Multi-Prototype Encoder for Object Recognition Using Multiple Prototype Images</title><author>Kang, Jun Seok ; Ahn, Sang Chul</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c358t-46fba21773555e917d629e8cfb46db09872929a7d98c8d9c9d00205e2956c4aa3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Coders</topic><topic>Deep learning</topic><topic>embedding space</topic><topic>Feature extraction</topic><topic>image classification</topic><topic>Labeling</topic><topic>Learning</topic><topic>Neural networks</topic><topic>Object recognition</topic><topic>Performance degradation</topic><topic>Perturbation methods</topic><topic>prototype learning</topic><topic>Prototypes</topic><topic>Prototyping</topic><topic>Task analysis</topic><topic>Training</topic><topic>variational encoder</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Kang, Jun Seok</creatorcontrib><creatorcontrib>Ahn, Sang Chul</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE Xplore Open Access Journals</collection><collection>IEEE All-Society Periodicals Package (ASPP) Online</collection><collection>IEEE Xplore</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Engineered Materials Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials 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>Directory of Open Access Journals</collection><jtitle>IEEE access</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kang, Jun Seok</au><au>Ahn, Sang Chul</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Variational Multi-Prototype Encoder for Object Recognition Using Multiple Prototype Images</atitle><jtitle>IEEE access</jtitle><stitle>Access</stitle><date>2022</date><risdate>2022</risdate><volume>10</volume><spage>19586</spage><epage>19598</epage><pages>19586-19598</pages><issn>2169-3536</issn><eissn>2169-3536</eissn><coden>IAECCG</coden><abstract>In the recent research of Variational Prototyping-Encoder (VPE), the problem of classifying 2D flat objects of the unseen class has been addressed. VPE solves this problem by pre-learning the image translation task from real-world object images to their corresponding prototype images as a meta-task. VPE uses a single prototype for each object class. However, in general, a single prototype is not sufficient to represent a generic object class because the appearance can change significantly according to viewpoints and other factors. In this case, using VPE and a single prototype for each class in training can result in overfitting or performance degradation. One solution may be the use of multiple prototypes. However, this also requires costly sub-labeling for dividing the input class into smaller classes and assigning a prototype to each. Therefore, we propose a new learning method, the variational multi-prototype encoder (VaMPE), which can overcome the limitations of VPE and use multiple prototypes for each object class. The proposed method does not require additional sub-labeling other than simply adding multiple prototypes to each class. Through various experiments, we demonstrate that the proposed method outperforms VPE.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/ACCESS.2022.3151856</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0002-6384-891X</orcidid><orcidid>https://orcid.org/0000-0003-2959-1089</orcidid><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 2169-3536 |
ispartof | IEEE access, 2022, Vol.10, p.19586-19598 |
issn | 2169-3536 2169-3536 |
language | eng |
recordid | cdi_proquest_journals_2633041944 |
source | Directory of Open Access Journals; Free E-Journal (出版社公開部分のみ); IEEE Xplore Open Access Journals |
subjects | Coders Deep learning embedding space Feature extraction image classification Labeling Learning Neural networks Object recognition Performance degradation Perturbation methods prototype learning Prototypes Prototyping Task analysis Training variational encoder |
title | Variational Multi-Prototype Encoder for Object Recognition Using Multiple Prototype Images |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T07%3A00%3A02IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_doaj_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Variational%20Multi-Prototype%20Encoder%20for%20Object%20Recognition%20Using%20Multiple%20Prototype%20Images&rft.jtitle=IEEE%20access&rft.au=Kang,%20Jun%20Seok&rft.date=2022&rft.volume=10&rft.spage=19586&rft.epage=19598&rft.pages=19586-19598&rft.issn=2169-3536&rft.eissn=2169-3536&rft.coden=IAECCG&rft_id=info:doi/10.1109/ACCESS.2022.3151856&rft_dat=%3Cproquest_doaj_%3E2633041944%3C/proquest_doaj_%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2633041944&rft_id=info:pmid/&rft_ieee_id=9714291&rft_doaj_id=oai_doaj_org_article_cdc0924d57624fa1b5e1a0813fd95b4b&rfr_iscdi=true |