Human Pose Tracking Using Multi-level Structured Models
Tracking body poses of multiple persons in monocular video is a challenging problem due to the high dimensionality of the state space and issues such as inter-occlusion of the persons’ bodies. We proposed a three-stage approach with a multi-level state representation that enables a hierarchical esti...
Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 381 |
|---|---|
| container_issue | |
| container_start_page | 368 |
| container_title | |
| container_volume | |
| creator | Lee, Mun Wai Nevatia, Ram |
| description | Tracking body poses of multiple persons in monocular video is a challenging problem due to the high dimensionality of the state space and issues such as inter-occlusion of the persons’ bodies. We proposed a three-stage approach with a multi-level state representation that enables a hierarchical estimation of 3D body poses. At the first stage, humans are tracked as blobs. In the second stage, parts such as face, shoulders and limbs are estimated and estimates are combined by grid-based belief propagation to infer 2D joint positions. The derived belief maps are used as proposal functions in the third stage to infer the 3D pose using data-driven Markov chain Monte Carlo. Experimental results on realistic indoor video sequences show that the method is able to track multiple persons during complex movement such as turning movement with inter-occlusion. |
| doi_str_mv | 10.1007/11744078_29 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>pascalfrancis_sprin</sourceid><recordid>TN_cdi_pascalfrancis_primary_20046224</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>20046224</sourcerecordid><originalsourceid>FETCH-LOGICAL-p219t-db3e21260d8ee27133ea974f0b2be00badd4c629cddb603e95e0c884f9aeb5bd3</originalsourceid><addsrcrecordid>eNpVUD1PwzAUNF8SVenEH8jCwBB49nPteEQVpUitQKKdIzt-qULTpLITJP49qcoAN9wNdzqdjrFbDg8cQD9yrqUEneXCnLGJ0RlOJSBmqNU5G3HFeYoozcU_T5lLNgIEkRot8ZpNYvyEAciV4WbE9KLf2yZ5byMl62CLXdVsk0088qqvuyqt6Yvq5KMLfdH1gXyyaj3V8YZdlbaONPnVMdvMn9ezRbp8e3mdPS3Tg-CmS71DElwo8BmR0ByR7DCkBCccATjrvSyUMIX3TgGSmRIUWSZLY8lNnccxuzv1HmwsbF0G2xRVzA-h2tvwnQsAqYSQQ-7-lIuD1Wwp5K5tdzHnkB_Py_-chz_KkFtR</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Human Pose Tracking Using Multi-level Structured Models</title><source>Springer Books</source><creator>Lee, Mun Wai ; Nevatia, Ram</creator><contributor>Bischof, Horst ; Pinz, Axel ; Leonardis, Aleš</contributor><creatorcontrib>Lee, Mun Wai ; Nevatia, Ram ; Bischof, Horst ; Pinz, Axel ; Leonardis, Aleš</creatorcontrib><description>Tracking body poses of multiple persons in monocular video is a challenging problem due to the high dimensionality of the state space and issues such as inter-occlusion of the persons’ bodies. We proposed a three-stage approach with a multi-level state representation that enables a hierarchical estimation of 3D body poses. At the first stage, humans are tracked as blobs. In the second stage, parts such as face, shoulders and limbs are estimated and estimates are combined by grid-based belief propagation to infer 2D joint positions. The derived belief maps are used as proposal functions in the third stage to infer the 3D pose using data-driven Markov chain Monte Carlo. Experimental results on realistic indoor video sequences show that the method is able to track multiple persons during complex movement such as turning movement with inter-occlusion.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 9783540338369</identifier><identifier>ISBN: 3540338365</identifier><identifier>ISBN: 9783540338321</identifier><identifier>ISBN: 3540338322</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 9783540338376</identifier><identifier>EISBN: 3540338373</identifier><identifier>DOI: 10.1007/11744078_29</identifier><language>eng</language><publisher>Berlin, Heidelberg: Springer Berlin Heidelberg</publisher><subject>Applied sciences ; Artificial intelligence ; Belief Propagation ; Body Joint ; Computer science; control theory; systems ; Exact sciences and technology ; Motion Capture Data ; Observation Function ; Pattern recognition. Digital image processing. Computational geometry ; State Candidate</subject><ispartof>Computer Vision – ECCV 2006, 2006, p.368-381</ispartof><rights>Springer-Verlag Berlin Heidelberg 2006</rights><rights>2008 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/11744078_29$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/11744078_29$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>309,310,779,780,784,789,790,793,4050,4051,27925,38255,41442,42511</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=20046224$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><contributor>Bischof, Horst</contributor><contributor>Pinz, Axel</contributor><contributor>Leonardis, Aleš</contributor><creatorcontrib>Lee, Mun Wai</creatorcontrib><creatorcontrib>Nevatia, Ram</creatorcontrib><title>Human Pose Tracking Using Multi-level Structured Models</title><title>Computer Vision – ECCV 2006</title><description>Tracking body poses of multiple persons in monocular video is a challenging problem due to the high dimensionality of the state space and issues such as inter-occlusion of the persons’ bodies. We proposed a three-stage approach with a multi-level state representation that enables a hierarchical estimation of 3D body poses. At the first stage, humans are tracked as blobs. In the second stage, parts such as face, shoulders and limbs are estimated and estimates are combined by grid-based belief propagation to infer 2D joint positions. The derived belief maps are used as proposal functions in the third stage to infer the 3D pose using data-driven Markov chain Monte Carlo. Experimental results on realistic indoor video sequences show that the method is able to track multiple persons during complex movement such as turning movement with inter-occlusion.</description><subject>Applied sciences</subject><subject>Artificial intelligence</subject><subject>Belief Propagation</subject><subject>Body Joint</subject><subject>Computer science; control theory; systems</subject><subject>Exact sciences and technology</subject><subject>Motion Capture Data</subject><subject>Observation Function</subject><subject>Pattern recognition. Digital image processing. Computational geometry</subject><subject>State Candidate</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>9783540338369</isbn><isbn>3540338365</isbn><isbn>9783540338321</isbn><isbn>3540338322</isbn><isbn>9783540338376</isbn><isbn>3540338373</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2006</creationdate><recordtype>conference_proceeding</recordtype><recordid>eNpVUD1PwzAUNF8SVenEH8jCwBB49nPteEQVpUitQKKdIzt-qULTpLITJP49qcoAN9wNdzqdjrFbDg8cQD9yrqUEneXCnLGJ0RlOJSBmqNU5G3HFeYoozcU_T5lLNgIEkRot8ZpNYvyEAciV4WbE9KLf2yZ5byMl62CLXdVsk0088qqvuyqt6Yvq5KMLfdH1gXyyaj3V8YZdlbaONPnVMdvMn9ezRbp8e3mdPS3Tg-CmS71DElwo8BmR0ByR7DCkBCccATjrvSyUMIX3TgGSmRIUWSZLY8lNnccxuzv1HmwsbF0G2xRVzA-h2tvwnQsAqYSQQ-7-lIuD1Wwp5K5tdzHnkB_Py_-chz_KkFtR</recordid><startdate>2006</startdate><enddate>2006</enddate><creator>Lee, Mun Wai</creator><creator>Nevatia, Ram</creator><general>Springer Berlin Heidelberg</general><general>Springer</general><scope>IQODW</scope></search><sort><creationdate>2006</creationdate><title>Human Pose Tracking Using Multi-level Structured Models</title><author>Lee, Mun Wai ; Nevatia, Ram</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p219t-db3e21260d8ee27133ea974f0b2be00badd4c629cddb603e95e0c884f9aeb5bd3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2006</creationdate><topic>Applied sciences</topic><topic>Artificial intelligence</topic><topic>Belief Propagation</topic><topic>Body Joint</topic><topic>Computer science; control theory; systems</topic><topic>Exact sciences and technology</topic><topic>Motion Capture Data</topic><topic>Observation Function</topic><topic>Pattern recognition. Digital image processing. Computational geometry</topic><topic>State Candidate</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Lee, Mun Wai</creatorcontrib><creatorcontrib>Nevatia, Ram</creatorcontrib><collection>Pascal-Francis</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Lee, Mun Wai</au><au>Nevatia, Ram</au><au>Bischof, Horst</au><au>Pinz, Axel</au><au>Leonardis, Aleš</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Human Pose Tracking Using Multi-level Structured Models</atitle><btitle>Computer Vision – ECCV 2006</btitle><date>2006</date><risdate>2006</risdate><spage>368</spage><epage>381</epage><pages>368-381</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>9783540338369</isbn><isbn>3540338365</isbn><isbn>9783540338321</isbn><isbn>3540338322</isbn><eisbn>9783540338376</eisbn><eisbn>3540338373</eisbn><abstract>Tracking body poses of multiple persons in monocular video is a challenging problem due to the high dimensionality of the state space and issues such as inter-occlusion of the persons’ bodies. We proposed a three-stage approach with a multi-level state representation that enables a hierarchical estimation of 3D body poses. At the first stage, humans are tracked as blobs. In the second stage, parts such as face, shoulders and limbs are estimated and estimates are combined by grid-based belief propagation to infer 2D joint positions. The derived belief maps are used as proposal functions in the third stage to infer the 3D pose using data-driven Markov chain Monte Carlo. Experimental results on realistic indoor video sequences show that the method is able to track multiple persons during complex movement such as turning movement with inter-occlusion.</abstract><cop>Berlin, Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/11744078_29</doi><tpages>14</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0302-9743 |
| ispartof | Computer Vision – ECCV 2006, 2006, p.368-381 |
| issn | 0302-9743 1611-3349 |
| language | eng |
| recordid | cdi_pascalfrancis_primary_20046224 |
| source | Springer Books |
| subjects | Applied sciences Artificial intelligence Belief Propagation Body Joint Computer science control theory systems Exact sciences and technology Motion Capture Data Observation Function Pattern recognition. Digital image processing. Computational geometry State Candidate |
| title | Human Pose Tracking Using Multi-level Structured Models |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-05T19%3A12%3A23IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-pascalfrancis_sprin&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Human%20Pose%20Tracking%20Using%20Multi-level%20Structured%20Models&rft.btitle=Computer%20Vision%20%E2%80%93%20ECCV%202006&rft.au=Lee,%20Mun%20Wai&rft.date=2006&rft.spage=368&rft.epage=381&rft.pages=368-381&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=9783540338369&rft.isbn_list=3540338365&rft.isbn_list=9783540338321&rft.isbn_list=3540338322&rft_id=info:doi/10.1007/11744078_29&rft_dat=%3Cpascalfrancis_sprin%3E20046224%3C/pascalfrancis_sprin%3E%3Curl%3E%3C/url%3E&rft.eisbn=9783540338376&rft.eisbn_list=3540338373&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |