No Cover Image

Journal article 638 views 185 downloads

From pose to activity: Surveying datasets and introducing CONVERSE / Michael Edwards, Jingjing Deng, Xianghua Xie

Computer Vision and Image Understanding, Volume: 144, Pages: 73 - 105

Swansea University Authors: Michael Edwards, Jingjing Deng, Xianghua Xie

Abstract

We present a review on the current state of publicly available datasets within the human action recognition community; highlighting the revival of pose based methods and recent progress of understanding person-person interaction modeling. We also propose a novel dataset that represents complex conve...

Full description

Published in: Computer Vision and Image Understanding
ISSN: 10773142
Published: 2016
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa26730
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2016-03-10T02:00:08Z
last_indexed 2019-04-15T08:47:43Z
id cronfa26730
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2019-04-09T16:41:38.1870192</datestamp><bib-version>v2</bib-version><id>26730</id><entry>2016-03-09</entry><title>From pose to activity: Surveying datasets and introducing CONVERSE</title><swanseaauthors><author><sid>684864a1ce01c3d774e83ed55e41770e</sid><ORCID>0000-0003-3367-969X</ORCID><firstname>Michael</firstname><surname>Edwards</surname><name>Michael Edwards</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>6f6d01d585363d6dc1622640bb4fcb3f</sid><ORCID>0000-0001-9274-651X</ORCID><firstname>Jingjing</firstname><surname>Deng</surname><name>Jingjing Deng</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>b334d40963c7a2f435f06d2c26c74e11</sid><ORCID>0000-0002-2701-8660</ORCID><firstname>Xianghua</firstname><surname>Xie</surname><name>Xianghua Xie</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2016-03-09</date><deptcode>SCS</deptcode><abstract>We present a review on the current state of publicly available datasets within the human action recognition community; highlighting the revival of pose based methods and recent progress of understanding person-person interaction modeling. We also propose a novel dataset that represents complex conversational interactions between two individuals via 3D pose. 8 pairwise interactions describing 7 separate conversation based scenarios were collected using two Kinect depth sensors. The intention is to provide events that are constructed from numerous primitive actions, interactions and motions, over a period of time; providing a set of subtle action classes that are more representative of the real world, and a chal- lenge to currently developed recognition methodologies. We believe this is among one of the first datasets devoted to conversational interaction classification using 3D pose features and the attributed papers show this task is indeed possible.</abstract><type>Journal Article</type><journal>Computer Vision and Image Understanding</journal><volume>144</volume><paginationStart>73</paginationStart><paginationEnd>105</paginationEnd><publisher/><issnPrint>10773142</issnPrint><keywords>Human Pose, Interaction, Action, Conversational Interaction, Data Set</keywords><publishedDay>31</publishedDay><publishedMonth>3</publishedMonth><publishedYear>2016</publishedYear><publishedDate>2016-03-31</publishedDate><doi>10.1016/j.cviu.2015.10.010</doi><url/><notes/><college>COLLEGE NANME</college><department>Computer Science</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>SCS</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2019-04-09T16:41:38.1870192</lastEdited><Created>2016-03-09T19:34:43.4985785</Created><path><level id="1">College of Science</level><level id="2">Computer Science</level></path><authors><author><firstname>Michael</firstname><surname>Edwards</surname><orcid>0000-0003-3367-969X</orcid><order>1</order></author><author><firstname>Michael</firstname><surname>Edwards</surname><order>2</order></author><author><firstname>Jingjing</firstname><surname>Deng</surname><orcid>0000-0001-9274-651X</orcid><order>3</order></author><author><firstname>Xianghua</firstname><surname>Xie</surname><orcid>0000-0002-2701-8660</orcid><order>4</order></author></authors><documents><document><filename>0026730-09032016193748.pdf</filename><originalFilename>CONVERSE.pdf</originalFilename><uploaded>2016-03-09T19:37:48.6170000</uploaded><type>Output</type><contentLength>2336765</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2016-03-09T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect></document></documents><OutputDurs/></rfc1807>
spelling 2019-04-09T16:41:38.1870192 v2 26730 2016-03-09 From pose to activity: Surveying datasets and introducing CONVERSE 684864a1ce01c3d774e83ed55e41770e 0000-0003-3367-969X Michael Edwards Michael Edwards true false 6f6d01d585363d6dc1622640bb4fcb3f 0000-0001-9274-651X Jingjing Deng Jingjing Deng true false b334d40963c7a2f435f06d2c26c74e11 0000-0002-2701-8660 Xianghua Xie Xianghua Xie true false 2016-03-09 SCS We present a review on the current state of publicly available datasets within the human action recognition community; highlighting the revival of pose based methods and recent progress of understanding person-person interaction modeling. We also propose a novel dataset that represents complex conversational interactions between two individuals via 3D pose. 8 pairwise interactions describing 7 separate conversation based scenarios were collected using two Kinect depth sensors. The intention is to provide events that are constructed from numerous primitive actions, interactions and motions, over a period of time; providing a set of subtle action classes that are more representative of the real world, and a chal- lenge to currently developed recognition methodologies. We believe this is among one of the first datasets devoted to conversational interaction classification using 3D pose features and the attributed papers show this task is indeed possible. Journal Article Computer Vision and Image Understanding 144 73 105 10773142 Human Pose, Interaction, Action, Conversational Interaction, Data Set 31 3 2016 2016-03-31 10.1016/j.cviu.2015.10.010 COLLEGE NANME Computer Science COLLEGE CODE SCS Swansea University 2019-04-09T16:41:38.1870192 2016-03-09T19:34:43.4985785 College of Science Computer Science Michael Edwards 0000-0003-3367-969X 1 Michael Edwards 2 Jingjing Deng 0000-0001-9274-651X 3 Xianghua Xie 0000-0002-2701-8660 4 0026730-09032016193748.pdf CONVERSE.pdf 2016-03-09T19:37:48.6170000 Output 2336765 application/pdf Accepted Manuscript true 2016-03-09T00:00:00.0000000 true
title From pose to activity: Surveying datasets and introducing CONVERSE
spellingShingle From pose to activity: Surveying datasets and introducing CONVERSE
Michael, Edwards
Jingjing, Deng
Xianghua, Xie
title_short From pose to activity: Surveying datasets and introducing CONVERSE
title_full From pose to activity: Surveying datasets and introducing CONVERSE
title_fullStr From pose to activity: Surveying datasets and introducing CONVERSE
title_full_unstemmed From pose to activity: Surveying datasets and introducing CONVERSE
title_sort From pose to activity: Surveying datasets and introducing CONVERSE
author_id_str_mv 684864a1ce01c3d774e83ed55e41770e
6f6d01d585363d6dc1622640bb4fcb3f
b334d40963c7a2f435f06d2c26c74e11
author_id_fullname_str_mv 684864a1ce01c3d774e83ed55e41770e_***_Michael, Edwards
6f6d01d585363d6dc1622640bb4fcb3f_***_Jingjing, Deng
b334d40963c7a2f435f06d2c26c74e11_***_Xianghua, Xie
author Michael, Edwards
Jingjing, Deng
Xianghua, Xie
author2 Michael Edwards
Michael Edwards
Jingjing Deng
Xianghua Xie
format Journal article
container_title Computer Vision and Image Understanding
container_volume 144
container_start_page 73
publishDate 2016
institution Swansea University
issn 10773142
doi_str_mv 10.1016/j.cviu.2015.10.010
college_str College of Science
hierarchytype
hierarchy_top_id collegeofscience
hierarchy_top_title College of Science
hierarchy_parent_id collegeofscience
hierarchy_parent_title College of Science
department_str Computer Science{{{_:::_}}}College of Science{{{_:::_}}}Computer Science
document_store_str 1
active_str 0
description We present a review on the current state of publicly available datasets within the human action recognition community; highlighting the revival of pose based methods and recent progress of understanding person-person interaction modeling. We also propose a novel dataset that represents complex conversational interactions between two individuals via 3D pose. 8 pairwise interactions describing 7 separate conversation based scenarios were collected using two Kinect depth sensors. The intention is to provide events that are constructed from numerous primitive actions, interactions and motions, over a period of time; providing a set of subtle action classes that are more representative of the real world, and a chal- lenge to currently developed recognition methodologies. We believe this is among one of the first datasets devoted to conversational interaction classification using 3D pose features and the attributed papers show this task is indeed possible.
published_date 2016-03-31T03:41:19Z
_version_ 1718366586718388224
score 10.843498