<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<site>sibgrapi.sid.inpe.br 802</site>
		<holdercode>{ibi 8JMKD3MGPEW34M/46T9EHH}</holdercode>
		<identifier>8JMKD3MGPEW34M/43B8A7P</identifier>
		<repository>sid.inpe.br/sibgrapi/2020/09.28.22.29</repository>
		<lastupdate>2020:09.28.22.29.43 sid.inpe.br/banon/2001/03.30.15.38 administrator</lastupdate>
		<metadatarepository>sid.inpe.br/sibgrapi/2020/09.28.22.29.43</metadatarepository>
		<metadatalastupdate>2022:06.14.00.00.10 sid.inpe.br/banon/2001/03.30.15.38 administrator {D 2020}</metadatalastupdate>
		<doi>10.1109/SIBGRAPI51738.2020.00051</doi>
		<citationkey>Schirmer:2020:Li2DPo</citationkey>
		<title>A lightweight 2D Pose Machine with attention enhancement</title>
		<format>On-line</format>
		<year>2020</year>
		<numberoffiles>1</numberoffiles>
		<size>7396 KiB</size>
		<author>Schirmer, Luiz,</author>
		<affiliation>PUC-rio</affiliation>
		<editor>Musse, Soraia Raupp,</editor>
		<editor>Cesar Junior, Roberto Marcondes,</editor>
		<editor>Pelechano, Nuria,</editor>
		<editor>Wang, Zhangyang (Atlas),</editor>
		<e-mailaddress>schirmer.luizj@gmail.com</e-mailaddress>
		<conferencename>Conference on Graphics, Patterns and Images, 33 (SIBGRAPI)</conferencename>
		<conferencelocation>Porto de Galinhas (virtual)</conferencelocation>
		<date>7-10 Nov. 2020</date>
		<publisher>IEEE Computer Society</publisher>
		<publisheraddress>Los Alamitos</publisheraddress>
		<booktitle>Proceedings</booktitle>
		<tertiarytype>Full Paper</tertiarytype>
		<transferableflag>1</transferableflag>
		<versiontype>finaldraft</versiontype>
		<keywords>pose estimation, tensor decompostion, attention layer.</keywords>
		<abstract>Pose  estimation  is  a  challenging  task  in  computer  vision that has many applications, as for example: in motion capture, in medical analysis, in human posture monitoring, and in robotics. In other words, it is a main tool to enable machines do understand human patterns in videos or images. Performing this task in real-time  while  maintaining  accuracy  and  precision  is  critical  for  many  of these applications. Several papers propose real time approaches considering deep neural networks for pose estimation. However, in most cases they fail when considering run-time performance or do not achieve the precision needed. In this work, we propose a new model for real-time pose estimation considering attention modules for convolutional neural networks (CNNs). We introduce a two-dimensional relative attention mechanism for feature  extraction  in  pose  machines  leading  to  improvements  in  accuracy. We create a single shot architecture where both operations to infer keypoints and part affinity fields share the information. Also, for each stage, we use tensor decompositions to not only reduce dimensionality, but also to improve performance. This allows us to factorize each convolution and drastically reduce the number of parameters in our network. Our experiments show that, with this factorized approach, it is possible to achieve state-of-art performance in terms of run-time while we have a small reduction in accuracy.</abstract>
		<language>en</language>
		<targetfile>Pose_estimation_for_Sibgrapi_2020.pdf</targetfile>
		<usergroup>schirmer.luizj@gmail.com</usergroup>
		<visibility>shown</visibility>
		<documentstage>not transferred</documentstage>
		<mirrorrepository>sid.inpe.br/banon/2001/03.30.15.38.24</mirrorrepository>
		<nexthigherunit>8JMKD3MGPEW34M/43G4L9S</nexthigherunit>
		<nexthigherunit>8JMKD3MGPEW34M/4742MCS</nexthigherunit>
		<citingitemlist>sid.inpe.br/sibgrapi/2020/10.28.20.46 8</citingitemlist>
		<hostcollection>sid.inpe.br/banon/2001/03.30.15.38</hostcollection>
		<username>schirmer.luizj@gmail.com</username>
		<agreement>agreement.html .htaccess .htaccess2</agreement>
		<lasthostcollection>sid.inpe.br/banon/2001/03.30.15.38</lasthostcollection>
		<url>http://sibgrapi.sid.inpe.br/rep-/sid.inpe.br/sibgrapi/2020/09.28.22.29</url>
	</metadata>
</metadatalist>