<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<site>sibgrapi.sid.inpe.br 802</site>
		<identifier>8JMKD3MGPBW34M/3EEH5B2</identifier>
		<repository>sid.inpe.br/sibgrapi/2013/07.11.04.39</repository>
		<lastupdate>2013:07.11.04.39.29 sid.inpe.br/banon/2001/03.30.15.38 virginiafernandesmota@gmail.com</lastupdate>
		<metadatarepository>sid.inpe.br/sibgrapi/2013/07.11.04.39.29</metadatarepository>
		<metadatalastupdate>2020:02.19.03.09.22 sid.inpe.br/banon/2001/03.30.15.38 administrator {D 2013}</metadatalastupdate>
		<citationkey>SadMotMacVieAra:2013:TeMoDe</citationkey>
		<title>A Tensor Motion Descriptor Based on Multiple Gradient Estimators</title>
		<format>On-line.</format>
		<year>2013</year>
		<date>Aug. 5-8, 2013</date>
		<numberoffiles>1</numberoffiles>
		<size>241 KiB</size>
		<author>Sad, Dhiego,</author>
		<author>Mota, Virgínia Fernandes,</author>
		<author>Maciel, Luiz Maurílio,</author>
		<author>Vieira, Marcelo Bernardes,</author>
		<author>Araújo, Arnaldo de Albuquerque,</author>
		<affiliation>Universidade Federal de Juiz de Fora</affiliation>
		<affiliation>Universidade Federal de Minas Gerais</affiliation>
		<affiliation>Universidade Federal de Juiz de Fora</affiliation>
		<affiliation>Universidade Federal de Juiz de Fora</affiliation>
		<affiliation>Universidade Federal de Minas Gerais</affiliation>
		<editor>Boyer, Kim,</editor>
		<editor>Hirata, Nina,</editor>
		<editor>Nedel, Luciana,</editor>
		<editor>Silva, Claudio,</editor>
		<e-mailaddress>virginiafernandesmota@gmail.com</e-mailaddress>
		<conferencename>Conference on Graphics, Patterns and Images, 26 (SIBGRAPI)</conferencename>
		<conferencelocation>Arequipa, Peru</conferencelocation>
		<booktitle>Proceedings</booktitle>
		<publisher>IEEE Computer Society</publisher>
		<publisheraddress>Los Alamitos</publisheraddress>
		<transferableflag>1</transferableflag>
		<contenttype>External Contribution</contenttype>
		<tertiarytype>Full Paper</tertiarytype>
		<keywords>Multifilter analysis, Motion descriptor, Orientation tensor, Human action recognition.</keywords>
		<abstract>This work presents a novel approach for motion description in videos using multiple band-pass filters which act as first order derivative estimators. The filters response on each frame are coded into individual histograms of gradients to reduce their dimensionality. They are combined using orientation tensors. No local features are extracted and no learning is performed, i.e., the descriptor depends uniquely on the input video. Motion description can be enhanced even using multiple filters with similar or overlapping frequency response. For the problem of human action recognition using the KTH database, our descriptor achieved the recognition rate of 93.3% using three Daubechies filters, one extra filter designed to correlate them, two-fold protocol and a SVM classifier. It is superior to most global descriptor approaches and fairly comparable to the state- of-the-art methods.</abstract>
		<language>en</language>
		<targetfile>paper_sad_114944.pdf</targetfile>
		<usergroup>virginiafernandesmota@gmail.com</usergroup>
		<visibility>shown</visibility>
		<mirrorrepository>sid.inpe.br/banon/2001/03.30.15.38.24</mirrorrepository>
		<hostcollection>sid.inpe.br/banon/2001/03.30.15.38</hostcollection>
		<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/2013/07.11.04.39</url>
	</metadata>
</metadatalist>