Loading…
Interactive Open-Ended Learning for 3D Object Recognition: An Approach and Experiments
3D object detection and recognition is increasingly used for manipulation and navigation tasks in service robots. It involves segmenting the objects present in a scene, estimating a feature descriptor for the object view and, finally, recognizing the object view by comparing it to the known object c...
Saved in:
Published in: | Journal of intelligent & robotic systems 2015-12, Vol.80 (3-4), p.537-553 |
---|---|
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-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3 |
---|---|
cites | cdi_FETCH-LOGICAL-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3 |
container_end_page | 553 |
container_issue | 3-4 |
container_start_page | 537 |
container_title | Journal of intelligent & robotic systems |
container_volume | 80 |
creator | Kasaei, S. Hamidreza Oliveira, Miguel Lim, Gi Hyun Seabra Lopes, Luís Tomé, Ana Maria |
description | 3D object detection and recognition is increasingly used for manipulation and navigation tasks in service robots. It involves segmenting the objects present in a scene, estimating a feature descriptor for the object view and, finally, recognizing the object view by comparing it to the known object categories. This paper presents an efficient approach capable of learning and recognizing object categories in an interactive and open-ended manner. In this paper, “open-ended” implies that the set of object categories to be learned is not known in advance. The training instances are extracted from on-line experiences of a robot, and thus become gradually available over time, rather than at the beginning of the learning process. This paper focuses on two state-of-the-art questions: (1) How to automatically detect, conceptualize and recognize objects in 3D scenes in an open-ended manner? (2) How to acquire and use high-level knowledge obtained from the interaction with human users, namely when they provide category labels, in order to improve the system performance? This approach starts with a pre-processing step to remove irrelevant data and prepare a suitable point cloud for the subsequent processing. Clustering is then applied to detect object candidates, and object views are described based on a 3D shape descriptor called spin-image. Finally, a nearest-neighbor classification rule is used to predict the categories of the detected objects. A leave-one-out cross validation algorithm is used to compute precision and recall, in a classical off-line evaluation setting, for different system parameters. Also, an on-line evaluation protocol is used to assess the performance of the system in an open-ended setting. Results show that the proposed system is able to interact with human users, learning new object categories continuously over time. |
doi_str_mv | 10.1007/s10846-015-0189-z |
format | article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_1770312874</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3852946431</sourcerecordid><originalsourceid>FETCH-LOGICAL-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3</originalsourceid><addsrcrecordid>eNp1kMFKAzEQhoMoWKsP4C3gxcvqJNlsEm9FqxYKBVGvYZudrVva7JpsRfv0ptSDCB6GuXz_z8xHyDmDKwagriMDnRcZMJlGm2x7QAZMKpFBDuaQDMBwlgE3xTE5iXEJAEZLMyCvE99jKF3ffCCddeizsa-wolMsg2_8gtZtoOKOzuZLdD19QtcufNM3rb-hI09HXRfa0r3R0ld0_NlhaNbo-3hKjupyFfHsZw_Jy_34-fYxm84eJrejaeZyLftM4FxpVBWDSgvuJCtE4dAwoRUzUsvKAa-NQKEcc8XczPNcMOlqbtA540oxJJf73nTG-wZjb9dNdLhalR7bTbRMKRCMa5Un9OIPumw3wafrEsW1BM0LmSi2p1xoYwxY2y69VIYvy8DuTNu9aZtM251pu00Zvs_ExPoFhl_N_4a-AdqDf-E</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1728508265</pqid></control><display><type>article</type><title>Interactive Open-Ended Learning for 3D Object Recognition: An Approach and Experiments</title><source>Springer Link</source><creator>Kasaei, S. Hamidreza ; Oliveira, Miguel ; Lim, Gi Hyun ; Seabra Lopes, Luís ; Tomé, Ana Maria</creator><creatorcontrib>Kasaei, S. Hamidreza ; Oliveira, Miguel ; Lim, Gi Hyun ; Seabra Lopes, Luís ; Tomé, Ana Maria</creatorcontrib><description>3D object detection and recognition is increasingly used for manipulation and navigation tasks in service robots. It involves segmenting the objects present in a scene, estimating a feature descriptor for the object view and, finally, recognizing the object view by comparing it to the known object categories. This paper presents an efficient approach capable of learning and recognizing object categories in an interactive and open-ended manner. In this paper, “open-ended” implies that the set of object categories to be learned is not known in advance. The training instances are extracted from on-line experiences of a robot, and thus become gradually available over time, rather than at the beginning of the learning process. This paper focuses on two state-of-the-art questions: (1) How to automatically detect, conceptualize and recognize objects in 3D scenes in an open-ended manner? (2) How to acquire and use high-level knowledge obtained from the interaction with human users, namely when they provide category labels, in order to improve the system performance? This approach starts with a pre-processing step to remove irrelevant data and prepare a suitable point cloud for the subsequent processing. Clustering is then applied to detect object candidates, and object views are described based on a 3D shape descriptor called spin-image. Finally, a nearest-neighbor classification rule is used to predict the categories of the detected objects. A leave-one-out cross validation algorithm is used to compute precision and recall, in a classical off-line evaluation setting, for different system parameters. Also, an on-line evaluation protocol is used to assess the performance of the system in an open-ended setting. Results show that the proposed system is able to interact with human users, learning new object categories continuously over time.</description><identifier>ISSN: 0921-0296</identifier><identifier>EISSN: 1573-0409</identifier><identifier>DOI: 10.1007/s10846-015-0189-z</identifier><language>eng</language><publisher>Dordrecht: Springer Netherlands</publisher><subject>Artificial Intelligence ; Categories ; Control ; Electrical Engineering ; Engineering ; Interactive ; Learning ; Mechanical Engineering ; Mechatronics ; Object recognition ; On-line systems ; Robotics ; Robots ; Three dimensional ; Three dimensional models</subject><ispartof>Journal of intelligent & robotic systems, 2015-12, Vol.80 (3-4), p.537-553</ispartof><rights>Springer Science+Business Media Dordrecht 2015</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3</citedby><cites>FETCH-LOGICAL-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Kasaei, S. Hamidreza</creatorcontrib><creatorcontrib>Oliveira, Miguel</creatorcontrib><creatorcontrib>Lim, Gi Hyun</creatorcontrib><creatorcontrib>Seabra Lopes, Luís</creatorcontrib><creatorcontrib>Tomé, Ana Maria</creatorcontrib><title>Interactive Open-Ended Learning for 3D Object Recognition: An Approach and Experiments</title><title>Journal of intelligent & robotic systems</title><addtitle>J Intell Robot Syst</addtitle><description>3D object detection and recognition is increasingly used for manipulation and navigation tasks in service robots. It involves segmenting the objects present in a scene, estimating a feature descriptor for the object view and, finally, recognizing the object view by comparing it to the known object categories. This paper presents an efficient approach capable of learning and recognizing object categories in an interactive and open-ended manner. In this paper, “open-ended” implies that the set of object categories to be learned is not known in advance. The training instances are extracted from on-line experiences of a robot, and thus become gradually available over time, rather than at the beginning of the learning process. This paper focuses on two state-of-the-art questions: (1) How to automatically detect, conceptualize and recognize objects in 3D scenes in an open-ended manner? (2) How to acquire and use high-level knowledge obtained from the interaction with human users, namely when they provide category labels, in order to improve the system performance? This approach starts with a pre-processing step to remove irrelevant data and prepare a suitable point cloud for the subsequent processing. Clustering is then applied to detect object candidates, and object views are described based on a 3D shape descriptor called spin-image. Finally, a nearest-neighbor classification rule is used to predict the categories of the detected objects. A leave-one-out cross validation algorithm is used to compute precision and recall, in a classical off-line evaluation setting, for different system parameters. Also, an on-line evaluation protocol is used to assess the performance of the system in an open-ended setting. Results show that the proposed system is able to interact with human users, learning new object categories continuously over time.</description><subject>Artificial Intelligence</subject><subject>Categories</subject><subject>Control</subject><subject>Electrical Engineering</subject><subject>Engineering</subject><subject>Interactive</subject><subject>Learning</subject><subject>Mechanical Engineering</subject><subject>Mechatronics</subject><subject>Object recognition</subject><subject>On-line systems</subject><subject>Robotics</subject><subject>Robots</subject><subject>Three dimensional</subject><subject>Three dimensional models</subject><issn>0921-0296</issn><issn>1573-0409</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2015</creationdate><recordtype>article</recordtype><recordid>eNp1kMFKAzEQhoMoWKsP4C3gxcvqJNlsEm9FqxYKBVGvYZudrVva7JpsRfv0ptSDCB6GuXz_z8xHyDmDKwagriMDnRcZMJlGm2x7QAZMKpFBDuaQDMBwlgE3xTE5iXEJAEZLMyCvE99jKF3ffCCddeizsa-wolMsg2_8gtZtoOKOzuZLdD19QtcufNM3rb-hI09HXRfa0r3R0ld0_NlhaNbo-3hKjupyFfHsZw_Jy_34-fYxm84eJrejaeZyLftM4FxpVBWDSgvuJCtE4dAwoRUzUsvKAa-NQKEcc8XczPNcMOlqbtA540oxJJf73nTG-wZjb9dNdLhalR7bTbRMKRCMa5Un9OIPumw3wafrEsW1BM0LmSi2p1xoYwxY2y69VIYvy8DuTNu9aZtM251pu00Zvs_ExPoFhl_N_4a-AdqDf-E</recordid><startdate>20151201</startdate><enddate>20151201</enddate><creator>Kasaei, S. Hamidreza</creator><creator>Oliveira, Miguel</creator><creator>Lim, Gi Hyun</creator><creator>Seabra Lopes, Luís</creator><creator>Tomé, Ana Maria</creator><general>Springer Netherlands</general><general>Springer Nature B.V</general><scope>AAYXX</scope><scope>CITATION</scope><scope>3V.</scope><scope>7SC</scope><scope>7SP</scope><scope>7TB</scope><scope>7XB</scope><scope>8AL</scope><scope>8FD</scope><scope>8FE</scope><scope>8FG</scope><scope>8FK</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>FR3</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>K7-</scope><scope>L6V</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>M0N</scope><scope>M7S</scope><scope>P5Z</scope><scope>P62</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>Q9U</scope><scope>F28</scope></search><sort><creationdate>20151201</creationdate><title>Interactive Open-Ended Learning for 3D Object Recognition: An Approach and Experiments</title><author>Kasaei, S. Hamidreza ; Oliveira, Miguel ; Lim, Gi Hyun ; Seabra Lopes, Luís ; Tomé, Ana Maria</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2015</creationdate><topic>Artificial Intelligence</topic><topic>Categories</topic><topic>Control</topic><topic>Electrical Engineering</topic><topic>Engineering</topic><topic>Interactive</topic><topic>Learning</topic><topic>Mechanical Engineering</topic><topic>Mechatronics</topic><topic>Object recognition</topic><topic>On-line systems</topic><topic>Robotics</topic><topic>Robots</topic><topic>Three dimensional</topic><topic>Three dimensional models</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Kasaei, S. Hamidreza</creatorcontrib><creatorcontrib>Oliveira, Miguel</creatorcontrib><creatorcontrib>Lim, Gi Hyun</creatorcontrib><creatorcontrib>Seabra Lopes, Luís</creatorcontrib><creatorcontrib>Tomé, Ana Maria</creatorcontrib><collection>CrossRef</collection><collection>ProQuest Central (Corporate)</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics & Communications Abstracts</collection><collection>Mechanical & Transportation Engineering Abstracts</collection><collection>ProQuest Central (purchase pre-March 2016)</collection><collection>Computing Database (Alumni Edition)</collection><collection>Technology Research Database</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Central (Alumni) (purchase pre-March 2016)</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>Advanced Technologies & Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>Engineering Research Database</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>Computer science database</collection><collection>ProQuest Engineering 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>Computing Database</collection><collection>Engineering Database</collection><collection>ProQuest advanced technologies & aerospace journals</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering collection</collection><collection>ProQuest Central Basic</collection><collection>ANTE: Abstracts in New Technology & Engineering</collection><jtitle>Journal of intelligent & robotic systems</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kasaei, S. Hamidreza</au><au>Oliveira, Miguel</au><au>Lim, Gi Hyun</au><au>Seabra Lopes, Luís</au><au>Tomé, Ana Maria</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Interactive Open-Ended Learning for 3D Object Recognition: An Approach and Experiments</atitle><jtitle>Journal of intelligent & robotic systems</jtitle><stitle>J Intell Robot Syst</stitle><date>2015-12-01</date><risdate>2015</risdate><volume>80</volume><issue>3-4</issue><spage>537</spage><epage>553</epage><pages>537-553</pages><issn>0921-0296</issn><eissn>1573-0409</eissn><abstract>3D object detection and recognition is increasingly used for manipulation and navigation tasks in service robots. It involves segmenting the objects present in a scene, estimating a feature descriptor for the object view and, finally, recognizing the object view by comparing it to the known object categories. This paper presents an efficient approach capable of learning and recognizing object categories in an interactive and open-ended manner. In this paper, “open-ended” implies that the set of object categories to be learned is not known in advance. The training instances are extracted from on-line experiences of a robot, and thus become gradually available over time, rather than at the beginning of the learning process. This paper focuses on two state-of-the-art questions: (1) How to automatically detect, conceptualize and recognize objects in 3D scenes in an open-ended manner? (2) How to acquire and use high-level knowledge obtained from the interaction with human users, namely when they provide category labels, in order to improve the system performance? This approach starts with a pre-processing step to remove irrelevant data and prepare a suitable point cloud for the subsequent processing. Clustering is then applied to detect object candidates, and object views are described based on a 3D shape descriptor called spin-image. Finally, a nearest-neighbor classification rule is used to predict the categories of the detected objects. A leave-one-out cross validation algorithm is used to compute precision and recall, in a classical off-line evaluation setting, for different system parameters. Also, an on-line evaluation protocol is used to assess the performance of the system in an open-ended setting. Results show that the proposed system is able to interact with human users, learning new object categories continuously over time.</abstract><cop>Dordrecht</cop><pub>Springer Netherlands</pub><doi>10.1007/s10846-015-0189-z</doi><tpages>17</tpages></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0921-0296 |
ispartof | Journal of intelligent & robotic systems, 2015-12, Vol.80 (3-4), p.537-553 |
issn | 0921-0296 1573-0409 |
language | eng |
recordid | cdi_proquest_miscellaneous_1770312874 |
source | Springer Link |
subjects | Artificial Intelligence Categories Control Electrical Engineering Engineering Interactive Learning Mechanical Engineering Mechatronics Object recognition On-line systems Robotics Robots Three dimensional Three dimensional models |
title | Interactive Open-Ended Learning for 3D Object Recognition: An Approach and Experiments |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T08%3A03%3A48IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Interactive%20Open-Ended%20Learning%20for%203D%20Object%20Recognition:%20An%20Approach%20and%20Experiments&rft.jtitle=Journal%20of%20intelligent%20&%20robotic%20systems&rft.au=Kasaei,%20S.%20Hamidreza&rft.date=2015-12-01&rft.volume=80&rft.issue=3-4&rft.spage=537&rft.epage=553&rft.pages=537-553&rft.issn=0921-0296&rft.eissn=1573-0409&rft_id=info:doi/10.1007/s10846-015-0189-z&rft_dat=%3Cproquest_cross%3E3852946431%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c485t-3eb78e7d10d832c51636ce9138719585dc02f93e37c1c6b9b44315cf29ecc9ca3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1728508265&rft_id=info:pmid/&rfr_iscdi=true |