<?xml version="1.0" encoding="ISO-8859-1"?>
<metadatalist>
	<metadata ReferenceType="Conference Proceedings">
		<identifier>6qtX3pFwXQZG2LgkFdY/LzdxM</identifier>
		<repository>sid.inpe.br/sibgrapi@80/2006/06.26.11.48</repository>
		<metadatarepository>sid.inpe.br/sibgrapi@80/2006/06.26.11.48.25</metadatarepository>
		<site>sibgrapi.sid.inpe.br 802</site>
		<citationkey>RibeiroGonz:2006:CoAn</citationkey>
		<author>Ribeiro, Hebert Luchetti,</author>
		<author>Gonzaga, Adilson,</author>
		<affiliation>School of Engineering at Sao Carlos</affiliation>
		<affiliation>School of Engineering at Sao Carlos</affiliation>
		<title>Hand Image Segmentation in Video Sequence by GMM: a comparative analysis</title>
		<conferencename>Brazilian Symposium on Computer Graphics and Image Processing, 19 (SIBGRAPI)</conferencename>
		<year>2006</year>
		<editor>Oliveira Neto, Manuel Menezes de,</editor>
		<editor>Carceroni, Rodrigo Lima,</editor>
		<booktitle>Proceedings</booktitle>
		<date>8-11 Oct. 2006</date>
		<publisheraddress>Los Alamitos</publisheraddress>
		<publisher>IEEE Computer Society</publisher>
		<conferencelocation>Manaus</conferencelocation>
		<keywords>Video Image Segmentation, Gaussian Mixture Model.</keywords>
		<abstract>This paper describes different approaches of realtimeGMM (Gaussian Mixture Method) backgroundsubtraction algorithm using video sequences for handimage segmentation. In each captured image, thesegmentation takes place where pixels belonging to thehands are separated from the background based onbackground extraction and skin-color segmentation. Atime-adaptive mixture of Gaussians is used to modelthe distribution of each pixel color value. For an inputimage, every new pixel value is checked, deciding if itmatches with one of the existing Gaussians based onthe distance from the mean in terms of the standarddeviation. The best matching distribution parametersare updated and its weight is increased. It is assumedthat the values of the background pixels have lowvariance and large weight. These matched pixels,considered as foreground, are compared based on skincolor thresholds. The hands position and otherattributes are tracked by frame. That enables us todistinguish the hand movement from the backgroundand other objects in movement, as well as to extractthe information from the movement for dynamic handgesture recognition.</abstract>
		<language>en</language>
		<tertiarytype>Full Paper</tertiarytype>
		<format>On-line</format>
		<size>1015 KiB</size>
		<numberoffiles>1</numberoffiles>
		<targetfile>Ribeiro-HandImageSegmentationInVideoSequenceByGMM.pdf</targetfile>
		<lastupdate>2006:07.10.17.21.57 sid.inpe.br/banon/2001/03.30.15.38 administrator</lastupdate>
		<metadatalastupdate>2020:02.19.03.17.33 sid.inpe.br/banon/2001/03.30.15.38 administrator {D 2006}</metadatalastupdate>
		<e-mailaddress>agonzaga@sc.usp.br</e-mailaddress>
		<usergroup>adilson administrator</usergroup>
		<visibility>shown</visibility>
		<transferableflag>1</transferableflag>
		<hostcollection>sid.inpe.br/banon/2001/03.30.15.38</hostcollection>
		<contenttype>External Contribution</contenttype>
		<lasthostcollection>sid.inpe.br/banon/2001/03.30.15.38</lasthostcollection>
		<url>http://sibgrapi.sid.inpe.br/rep-/sid.inpe.br/sibgrapi@80/2006/06.26.11.48</url>
	</metadata>
</metadatalist>