Loading…
Continual Learning Strategy in One-Stage Object Detection Framework Based on Experience Replay for Autonomous Driving Vehicle
Object detection is an important aspect for autonomous driving vehicles (ADV), which may comprise of a machine learning model that detects a range of classes. As the deployment of ADV widens globally, the variety of objects to be detected may increase beyond the designated range of classes. Continua...
Saved in:
| Published in: | Sensors (Basel, Switzerland) Switzerland), 2020-11, Vol.20 (23), p.6777 |
|---|---|
| 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-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233 |
|---|---|
| cites | cdi_FETCH-LOGICAL-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233 |
| container_end_page | |
| container_issue | 23 |
| container_start_page | 6777 |
| container_title | Sensors (Basel, Switzerland) |
| container_volume | 20 |
| creator | Shieh, Jeng-Lun Haq, Qazi Mazhar Ul Haq, Muhamad Amirul Karam, Said Chondro, Peter Gao, De-Qin Ruan, Shanq-Jang |
| description | Object detection is an important aspect for autonomous driving vehicles (ADV), which may comprise of a machine learning model that detects a range of classes. As the deployment of ADV widens globally, the variety of objects to be detected may increase beyond the designated range of classes. Continual learning for object detection essentially ensure a robust adaptation of a model to detect additional classes on the fly. This study proposes a novel continual learning method for object detection that learns new object class(es) along with cumulative memory of classes from prior learning rounds to avoid any catastrophic forgetting. The results of PASCAL VOC 2007 have suggested that the proposed ER method obtains 4.3% of mAP drop compared against the all-classes learning, which is the lowest amongst other prior arts. |
| doi_str_mv | 10.3390/s20236777 |
| format | article |
| fullrecord | <record><control><sourceid>proquest_doaj_</sourceid><recordid>TN_cdi_doaj_primary_oai_doaj_org_article_dc8ebc54cda747738721eaecb43e1970</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><doaj_id>oai_doaj_org_article_dc8ebc54cda747738721eaecb43e1970</doaj_id><sourcerecordid>2465954140</sourcerecordid><originalsourceid>FETCH-LOGICAL-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233</originalsourceid><addsrcrecordid>eNpdkk1v1DAQhiMEoqVw4A8gS1zgEPBX7PiCVLYtVFppJQpcLceZpF6y9tZxCnvgv-OwZdVyGst-9Mzr0RTFS4LfMabw-5FiyoSU8lFxTDjlZU0pfnzvfFQ8G8c1zhRj9dPiiDEqcC34cfF7EXxyfjIDWoKJ3vkeXaVoEvQ75DxaeSivkukBrZo12ITOIOXigkcX0WzgZ4g_0EczQovy1fmvLUQH3gL6AtvB7FAXIjqdUvBhE6YRnUV3O7f4DtfODvC8eNKZYYQXd_Wk-HZx_nXxuVyuPl0uTpel5UKlkjWdVTXmUjZYcForaFtWS8BKVgpsxRqsRNsKqbqGYJBU1pYJRVhTE1HlT58Ul3tvG8xab6PbmLjTwTj99yLEXpuY5kS6tTU0tuK2NTI3zG0oAQO24QyIkji7Puxd26nZQGvB53END6QPX7y71n241VmGJeFZ8OZOEMPNBGPSGzdaGAbjIc9IUy4EVVQKkdHX_6HrMEWfRzVTlao44XOit3vKxjCOEbpDGIL1vCD6sCCZfXU__YH8txHsDzAHtf0</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2465954140</pqid></control><display><type>article</type><title>Continual Learning Strategy in One-Stage Object Detection Framework Based on Experience Replay for Autonomous Driving Vehicle</title><source>Publicly Available Content Database</source><source>PubMed Central</source><creator>Shieh, Jeng-Lun ; Haq, Qazi Mazhar Ul ; Haq, Muhamad Amirul ; Karam, Said ; Chondro, Peter ; Gao, De-Qin ; Ruan, Shanq-Jang</creator><creatorcontrib>Shieh, Jeng-Lun ; Haq, Qazi Mazhar Ul ; Haq, Muhamad Amirul ; Karam, Said ; Chondro, Peter ; Gao, De-Qin ; Ruan, Shanq-Jang</creatorcontrib><description>Object detection is an important aspect for autonomous driving vehicles (ADV), which may comprise of a machine learning model that detects a range of classes. As the deployment of ADV widens globally, the variety of objects to be detected may increase beyond the designated range of classes. Continual learning for object detection essentially ensure a robust adaptation of a model to detect additional classes on the fly. This study proposes a novel continual learning method for object detection that learns new object class(es) along with cumulative memory of classes from prior learning rounds to avoid any catastrophic forgetting. The results of PASCAL VOC 2007 have suggested that the proposed ER method obtains 4.3% of mAP drop compared against the all-classes learning, which is the lowest amongst other prior arts.</description><identifier>ISSN: 1424-8220</identifier><identifier>EISSN: 1424-8220</identifier><identifier>DOI: 10.3390/s20236777</identifier><identifier>PMID: 33260864</identifier><language>eng</language><publisher>Switzerland: MDPI AG</publisher><subject>autonomous driving vehicles ; Classification ; continual learning ; Datasets ; Deep learning ; Knowledge ; Neural networks ; one-stage object detection ; Proposals ; Vehicles</subject><ispartof>Sensors (Basel, Switzerland), 2020-11, Vol.20 (23), p.6777</ispartof><rights>2020. This work is licensed under http://creativecommons.org/licenses/by/3.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>2020 by the authors. 2020</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233</citedby><cites>FETCH-LOGICAL-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233</cites><orcidid>0000-0003-1516-0229 ; 0000-0002-9770-026X ; 0000-0002-8129-4623 ; 0000-0002-5198-0550 ; 0000-0002-1302-7689 ; 0000-0003-1075-8512</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.proquest.com/docview/2465954140/fulltextPDF?pq-origsite=primo$$EPDF$$P50$$Gproquest$$Hfree_for_read</linktopdf><linktohtml>$$Uhttps://www.proquest.com/docview/2465954140?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>230,314,723,776,780,881,25731,27901,27902,36989,36990,44566,53766,53768,74869</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/33260864$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Shieh, Jeng-Lun</creatorcontrib><creatorcontrib>Haq, Qazi Mazhar Ul</creatorcontrib><creatorcontrib>Haq, Muhamad Amirul</creatorcontrib><creatorcontrib>Karam, Said</creatorcontrib><creatorcontrib>Chondro, Peter</creatorcontrib><creatorcontrib>Gao, De-Qin</creatorcontrib><creatorcontrib>Ruan, Shanq-Jang</creatorcontrib><title>Continual Learning Strategy in One-Stage Object Detection Framework Based on Experience Replay for Autonomous Driving Vehicle</title><title>Sensors (Basel, Switzerland)</title><addtitle>Sensors (Basel)</addtitle><description>Object detection is an important aspect for autonomous driving vehicles (ADV), which may comprise of a machine learning model that detects a range of classes. As the deployment of ADV widens globally, the variety of objects to be detected may increase beyond the designated range of classes. Continual learning for object detection essentially ensure a robust adaptation of a model to detect additional classes on the fly. This study proposes a novel continual learning method for object detection that learns new object class(es) along with cumulative memory of classes from prior learning rounds to avoid any catastrophic forgetting. The results of PASCAL VOC 2007 have suggested that the proposed ER method obtains 4.3% of mAP drop compared against the all-classes learning, which is the lowest amongst other prior arts.</description><subject>autonomous driving vehicles</subject><subject>Classification</subject><subject>continual learning</subject><subject>Datasets</subject><subject>Deep learning</subject><subject>Knowledge</subject><subject>Neural networks</subject><subject>one-stage object detection</subject><subject>Proposals</subject><subject>Vehicles</subject><issn>1424-8220</issn><issn>1424-8220</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><sourceid>DOA</sourceid><recordid>eNpdkk1v1DAQhiMEoqVw4A8gS1zgEPBX7PiCVLYtVFppJQpcLceZpF6y9tZxCnvgv-OwZdVyGst-9Mzr0RTFS4LfMabw-5FiyoSU8lFxTDjlZU0pfnzvfFQ8G8c1zhRj9dPiiDEqcC34cfF7EXxyfjIDWoKJ3vkeXaVoEvQ75DxaeSivkukBrZo12ITOIOXigkcX0WzgZ4g_0EczQovy1fmvLUQH3gL6AtvB7FAXIjqdUvBhE6YRnUV3O7f4DtfODvC8eNKZYYQXd_Wk-HZx_nXxuVyuPl0uTpel5UKlkjWdVTXmUjZYcForaFtWS8BKVgpsxRqsRNsKqbqGYJBU1pYJRVhTE1HlT58Ul3tvG8xab6PbmLjTwTj99yLEXpuY5kS6tTU0tuK2NTI3zG0oAQO24QyIkji7Puxd26nZQGvB53END6QPX7y71n241VmGJeFZ8OZOEMPNBGPSGzdaGAbjIc9IUy4EVVQKkdHX_6HrMEWfRzVTlao44XOit3vKxjCOEbpDGIL1vCD6sCCZfXU__YH8txHsDzAHtf0</recordid><startdate>20201127</startdate><enddate>20201127</enddate><creator>Shieh, Jeng-Lun</creator><creator>Haq, Qazi Mazhar Ul</creator><creator>Haq, Muhamad Amirul</creator><creator>Karam, Said</creator><creator>Chondro, Peter</creator><creator>Gao, De-Qin</creator><creator>Ruan, Shanq-Jang</creator><general>MDPI AG</general><general>MDPI</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>3V.</scope><scope>7X7</scope><scope>7XB</scope><scope>88E</scope><scope>8FI</scope><scope>8FJ</scope><scope>8FK</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>FYUFA</scope><scope>GHDGH</scope><scope>K9.</scope><scope>M0S</scope><scope>M1P</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>7X8</scope><scope>5PM</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0003-1516-0229</orcidid><orcidid>https://orcid.org/0000-0002-9770-026X</orcidid><orcidid>https://orcid.org/0000-0002-8129-4623</orcidid><orcidid>https://orcid.org/0000-0002-5198-0550</orcidid><orcidid>https://orcid.org/0000-0002-1302-7689</orcidid><orcidid>https://orcid.org/0000-0003-1075-8512</orcidid></search><sort><creationdate>20201127</creationdate><title>Continual Learning Strategy in One-Stage Object Detection Framework Based on Experience Replay for Autonomous Driving Vehicle</title><author>Shieh, Jeng-Lun ; Haq, Qazi Mazhar Ul ; Haq, Muhamad Amirul ; Karam, Said ; Chondro, Peter ; Gao, De-Qin ; Ruan, Shanq-Jang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>autonomous driving vehicles</topic><topic>Classification</topic><topic>continual learning</topic><topic>Datasets</topic><topic>Deep learning</topic><topic>Knowledge</topic><topic>Neural networks</topic><topic>one-stage object detection</topic><topic>Proposals</topic><topic>Vehicles</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shieh, Jeng-Lun</creatorcontrib><creatorcontrib>Haq, Qazi Mazhar Ul</creatorcontrib><creatorcontrib>Haq, Muhamad Amirul</creatorcontrib><creatorcontrib>Karam, Said</creatorcontrib><creatorcontrib>Chondro, Peter</creatorcontrib><creatorcontrib>Gao, De-Qin</creatorcontrib><creatorcontrib>Ruan, Shanq-Jang</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>Health & Medical Collection</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Medical Database (Alumni Edition)</collection><collection>Hospital Premium Collection</collection><collection>Hospital Premium Collection (Alumni Edition)</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>Health Research Premium Collection</collection><collection>Health Research Premium Collection (Alumni)</collection><collection>ProQuest Health & Medical Complete (Alumni)</collection><collection>Health & Medical Collection (Alumni Edition)</collection><collection>Medical Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>MEDLINE - Academic</collection><collection>PubMed Central (Full Participant titles)</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>Sensors (Basel, Switzerland)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shieh, Jeng-Lun</au><au>Haq, Qazi Mazhar Ul</au><au>Haq, Muhamad Amirul</au><au>Karam, Said</au><au>Chondro, Peter</au><au>Gao, De-Qin</au><au>Ruan, Shanq-Jang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Continual Learning Strategy in One-Stage Object Detection Framework Based on Experience Replay for Autonomous Driving Vehicle</atitle><jtitle>Sensors (Basel, Switzerland)</jtitle><addtitle>Sensors (Basel)</addtitle><date>2020-11-27</date><risdate>2020</risdate><volume>20</volume><issue>23</issue><spage>6777</spage><pages>6777-</pages><issn>1424-8220</issn><eissn>1424-8220</eissn><abstract>Object detection is an important aspect for autonomous driving vehicles (ADV), which may comprise of a machine learning model that detects a range of classes. As the deployment of ADV widens globally, the variety of objects to be detected may increase beyond the designated range of classes. Continual learning for object detection essentially ensure a robust adaptation of a model to detect additional classes on the fly. This study proposes a novel continual learning method for object detection that learns new object class(es) along with cumulative memory of classes from prior learning rounds to avoid any catastrophic forgetting. The results of PASCAL VOC 2007 have suggested that the proposed ER method obtains 4.3% of mAP drop compared against the all-classes learning, which is the lowest amongst other prior arts.</abstract><cop>Switzerland</cop><pub>MDPI AG</pub><pmid>33260864</pmid><doi>10.3390/s20236777</doi><orcidid>https://orcid.org/0000-0003-1516-0229</orcidid><orcidid>https://orcid.org/0000-0002-9770-026X</orcidid><orcidid>https://orcid.org/0000-0002-8129-4623</orcidid><orcidid>https://orcid.org/0000-0002-5198-0550</orcidid><orcidid>https://orcid.org/0000-0002-1302-7689</orcidid><orcidid>https://orcid.org/0000-0003-1075-8512</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1424-8220 |
| ispartof | Sensors (Basel, Switzerland), 2020-11, Vol.20 (23), p.6777 |
| issn | 1424-8220 1424-8220 |
| language | eng |
| recordid | cdi_doaj_primary_oai_doaj_org_article_dc8ebc54cda747738721eaecb43e1970 |
| source | Publicly Available Content Database; PubMed Central |
| subjects | autonomous driving vehicles Classification continual learning Datasets Deep learning Knowledge Neural networks one-stage object detection Proposals Vehicles |
| title | Continual Learning Strategy in One-Stage Object Detection Framework Based on Experience Replay for Autonomous Driving Vehicle |
| url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-31T23%3A14%3A28IST&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=Continual%20Learning%20Strategy%20in%20One-Stage%20Object%20Detection%20Framework%20Based%20on%20Experience%20Replay%20for%20Autonomous%20Driving%20Vehicle&rft.jtitle=Sensors%20(Basel,%20Switzerland)&rft.au=Shieh,%20Jeng-Lun&rft.date=2020-11-27&rft.volume=20&rft.issue=23&rft.spage=6777&rft.pages=6777-&rft.issn=1424-8220&rft.eissn=1424-8220&rft_id=info:doi/10.3390/s20236777&rft_dat=%3Cproquest_doaj_%3E2465954140%3C/proquest_doaj_%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c469t-3bfc980477b064289edd387e09759ec53b096dd679fb10e7278c36913b8165233%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2465954140&rft_id=info:pmid/33260864&rfr_iscdi=true |