HUMBI: A Large Multiview Dataset of Human Body Expressions

This paper presents a new large multiview dataset called HUMBI for human body expressions with natural clothing. The goal of HUMBI is to facilitate modeling view-specific appearance and geometry of gaze, face, hand, body, and garment from assorted people. 107 synchronized HD cam- eras are used to ca...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu, Zhixuan, Shin Yoon, Jae, Lee, In Kyu, Venkatesh, Prashanth, Park, Jaesik, Yu, Jihun, Park, Hyun Soo
Format: Conference Proceeding
Language:English
Subjects:
Biological system modeling
Cameras
Face
Geometry
Solid modeling
Three-dimensional displays
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 2997
container_issue
container_start_page 2987
container_title
container_volume
creator Yu, Zhixuan
Shin Yoon, Jae
Lee, In Kyu
Venkatesh, Prashanth
Park, Jaesik
Yu, Jihun
Park, Hyun Soo
description This paper presents a new large multiview dataset called HUMBI for human body expressions with natural clothing. The goal of HUMBI is to facilitate modeling view-specific appearance and geometry of gaze, face, hand, body, and garment from assorted people. 107 synchronized HD cam- eras are used to capture 772 distinctive subjects across gen- der, ethnicity, age, and physical condition. With the mul- tiview image streams, we reconstruct high fidelity body ex- pressions using 3D mesh models, which allows representing view-specific appearance using their canonical atlas. We demonstrate that HUMBI is highly effective in learning and reconstructing a complete human model and is complemen- tary to the existing datasets of human body expressions with limited views and subjects such as MPII-Gaze, Multi-PIE, Human3.6M, and Panoptic Studio datasets.
doi_str_mv 10.1109/CVPR42600.2020.00306
format conference_proceeding
fullrecord <record><control><sourceid>ieee_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_9156401</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>9156401</ieee_id><sourcerecordid>9156401</sourcerecordid><originalsourceid>FETCH-LOGICAL-i203t-2555778c03d9403349c5b8c5b483e0403dc56e87e084ba62c955c82ae1b6f7033</originalsourceid><addsrcrecordid>eNotjM1Kw0AUhUdBsNQ8gS7mBRLv_M9018ZqCimKWLdlktzISNuUTKr27Q3o4uPA4TuHkDsGGWPg7vP3l1fJNUDGgUMGIEBfkMQZywwfYdqqSzLhyqjUgFHXJInxE0aPM6adnZBZsVkvVjM6p6XvP5CuT7shfAX8pg9-8BEH2rW0OO39gS665kyXP8ceYwzdId6Qq9bvIib_OSWbx-VbXqTl89Mqn5dp4CCGlCuljLE1iMZJEEK6WlV2RFqBMDZNrTRag2Bl5TWvnVK15R5ZpVszDqbk9u83IOL22Ie9789bx5SWwMQvzZ5GHg</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>HUMBI: A Large Multiview Dataset of Human Body Expressions</title><source>IEEE Xplore All Conference Series</source><creator>Yu, Zhixuan ; Shin Yoon, Jae ; Lee, In Kyu ; Venkatesh, Prashanth ; Park, Jaesik ; Yu, Jihun ; Park, Hyun Soo</creator><creatorcontrib>Yu, Zhixuan ; Shin Yoon, Jae ; Lee, In Kyu ; Venkatesh, Prashanth ; Park, Jaesik ; Yu, Jihun ; Park, Hyun Soo</creatorcontrib><description>This paper presents a new large multiview dataset called HUMBI for human body expressions with natural clothing. The goal of HUMBI is to facilitate modeling view-specific appearance and geometry of gaze, face, hand, body, and garment from assorted people. 107 synchronized HD cam- eras are used to capture 772 distinctive subjects across gen- der, ethnicity, age, and physical condition. With the mul- tiview image streams, we reconstruct high fidelity body ex- pressions using 3D mesh models, which allows representing view-specific appearance using their canonical atlas. We demonstrate that HUMBI is highly effective in learning and reconstructing a complete human model and is complemen- tary to the existing datasets of human body expressions with limited views and subjects such as MPII-Gaze, Multi-PIE, Human3.6M, and Panoptic Studio datasets.</description><identifier>EISSN: 2575-7075</identifier><identifier>EISBN: 9781728171685</identifier><identifier>EISBN: 1728171687</identifier><identifier>DOI: 10.1109/CVPR42600.2020.00306</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Biological system modeling ; Cameras ; Face ; Geometry ; Solid modeling ; Three-dimensional displays</subject><ispartof>2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, p.2987-2997</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/9156401$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,27904,54533,54910</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/9156401$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Yu, Zhixuan</creatorcontrib><creatorcontrib>Shin Yoon, Jae</creatorcontrib><creatorcontrib>Lee, In Kyu</creatorcontrib><creatorcontrib>Venkatesh, Prashanth</creatorcontrib><creatorcontrib>Park, Jaesik</creatorcontrib><creatorcontrib>Yu, Jihun</creatorcontrib><creatorcontrib>Park, Hyun Soo</creatorcontrib><title>HUMBI: A Large Multiview Dataset of Human Body Expressions</title><title>2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)</title><addtitle>CVPR</addtitle><description>This paper presents a new large multiview dataset called HUMBI for human body expressions with natural clothing. The goal of HUMBI is to facilitate modeling view-specific appearance and geometry of gaze, face, hand, body, and garment from assorted people. 107 synchronized HD cam- eras are used to capture 772 distinctive subjects across gen- der, ethnicity, age, and physical condition. With the mul- tiview image streams, we reconstruct high fidelity body ex- pressions using 3D mesh models, which allows representing view-specific appearance using their canonical atlas. We demonstrate that HUMBI is highly effective in learning and reconstructing a complete human model and is complemen- tary to the existing datasets of human body expressions with limited views and subjects such as MPII-Gaze, Multi-PIE, Human3.6M, and Panoptic Studio datasets.</description><subject>Biological system modeling</subject><subject>Cameras</subject><subject>Face</subject><subject>Geometry</subject><subject>Solid modeling</subject><subject>Three-dimensional displays</subject><issn>2575-7075</issn><isbn>9781728171685</isbn><isbn>1728171687</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2020</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotjM1Kw0AUhUdBsNQ8gS7mBRLv_M9018ZqCimKWLdlktzISNuUTKr27Q3o4uPA4TuHkDsGGWPg7vP3l1fJNUDGgUMGIEBfkMQZywwfYdqqSzLhyqjUgFHXJInxE0aPM6adnZBZsVkvVjM6p6XvP5CuT7shfAX8pg9-8BEH2rW0OO39gS665kyXP8ceYwzdId6Qq9bvIib_OSWbx-VbXqTl89Mqn5dp4CCGlCuljLE1iMZJEEK6WlV2RFqBMDZNrTRag2Bl5TWvnVK15R5ZpVszDqbk9u83IOL22Ie9789bx5SWwMQvzZ5GHg</recordid><startdate>202006</startdate><enddate>202006</enddate><creator>Yu, Zhixuan</creator><creator>Shin Yoon, Jae</creator><creator>Lee, In Kyu</creator><creator>Venkatesh, Prashanth</creator><creator>Park, Jaesik</creator><creator>Yu, Jihun</creator><creator>Park, Hyun Soo</creator><general>IEEE</general><scope>6IE</scope><scope>6IH</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIO</scope></search><sort><creationdate>202006</creationdate><title>HUMBI: A Large Multiview Dataset of Human Body Expressions</title><author>Yu, Zhixuan ; Shin Yoon, Jae ; Lee, In Kyu ; Venkatesh, Prashanth ; Park, Jaesik ; Yu, Jihun ; Park, Hyun Soo</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i203t-2555778c03d9403349c5b8c5b483e0403dc56e87e084ba62c955c82ae1b6f7033</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Biological system modeling</topic><topic>Cameras</topic><topic>Face</topic><topic>Geometry</topic><topic>Solid modeling</topic><topic>Three-dimensional displays</topic><toplevel>online_resources</toplevel><creatorcontrib>Yu, Zhixuan</creatorcontrib><creatorcontrib>Shin Yoon, Jae</creatorcontrib><creatorcontrib>Lee, In Kyu</creatorcontrib><creatorcontrib>Venkatesh, Prashanth</creatorcontrib><creatorcontrib>Park, Jaesik</creatorcontrib><creatorcontrib>Yu, Jihun</creatorcontrib><creatorcontrib>Park, Hyun Soo</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan (POP) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE/IET Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP) 1998-present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Yu, Zhixuan</au><au>Shin Yoon, Jae</au><au>Lee, In Kyu</au><au>Venkatesh, Prashanth</au><au>Park, Jaesik</au><au>Yu, Jihun</au><au>Park, Hyun Soo</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>HUMBI: A Large Multiview Dataset of Human Body Expressions</atitle><btitle>2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)</btitle><stitle>CVPR</stitle><date>2020-06</date><risdate>2020</risdate><spage>2987</spage><epage>2997</epage><pages>2987-2997</pages><eissn>2575-7075</eissn><eisbn>9781728171685</eisbn><eisbn>1728171687</eisbn><coden>IEEPAD</coden><abstract>This paper presents a new large multiview dataset called HUMBI for human body expressions with natural clothing. The goal of HUMBI is to facilitate modeling view-specific appearance and geometry of gaze, face, hand, body, and garment from assorted people. 107 synchronized HD cam- eras are used to capture 772 distinctive subjects across gen- der, ethnicity, age, and physical condition. With the mul- tiview image streams, we reconstruct high fidelity body ex- pressions using 3D mesh models, which allows representing view-specific appearance using their canonical atlas. We demonstrate that HUMBI is highly effective in learning and reconstructing a complete human model and is complemen- tary to the existing datasets of human body expressions with limited views and subjects such as MPII-Gaze, Multi-PIE, Human3.6M, and Panoptic Studio datasets.</abstract><pub>IEEE</pub><doi>10.1109/CVPR42600.2020.00306</doi><tpages>11</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier EISSN: 2575-7075
ispartof 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, p.2987-2997
issn 2575-7075
language eng
recordid cdi_ieee_primary_9156401
source IEEE Xplore All Conference Series
subjects Biological system modeling
Cameras
Face
Geometry
Solid modeling
Three-dimensional displays
title HUMBI: A Large Multiview Dataset of Human Body Expressions
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-26T07%3A17%3A37IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=HUMBI:%20A%20Large%20Multiview%20Dataset%20of%20Human%20Body%20Expressions&rft.btitle=2020%20IEEE/CVF%20Conference%20on%20Computer%20Vision%20and%20Pattern%20Recognition%20(CVPR)&rft.au=Yu,%20Zhixuan&rft.date=2020-06&rft.spage=2987&rft.epage=2997&rft.pages=2987-2997&rft.eissn=2575-7075&rft.coden=IEEPAD&rft_id=info:doi/10.1109/CVPR42600.2020.00306&rft.eisbn=9781728171685&rft.eisbn_list=1728171687&rft_dat=%3Cieee_CHZPO%3E9156401%3C/ieee_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i203t-2555778c03d9403349c5b8c5b483e0403dc56e87e084ba62c955c82ae1b6f7033%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=9156401&rfr_iscdi=true