<?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>8JMKD3MGPAW/3M5J6KE</identifier>
		<repository>sid.inpe.br/sibgrapi/2016/07.22.14.26</repository>
		<lastupdate>2016:07.22.14.37.43 sid.inpe.br/banon/2001/03.30.15.38 administrator</lastupdate>
		<metadatarepository>sid.inpe.br/sibgrapi/2016/07.22.14.26.25</metadatarepository>
		<metadatalastupdate>2022:06.14.00.08.34 sid.inpe.br/banon/2001/03.30.15.38 administrator {D 2016}</metadatalastupdate>
		<doi>10.1109/SIBGRAPI.2016.060</doi>
		<citationkey>MontagnerJrHiraCanu:2016:KeApWo</citationkey>
		<title>Kernel approximations for W-operator learning</title>
		<format>On-line</format>
		<year>2016</year>
		<numberoffiles>1</numberoffiles>
		<size>753 KiB</size>
		<author>Montagner, Igor S.,</author>
		<author>Jr., Roberto Hirata,</author>
		<author>Hirata, Nina S. T.,</author>
		<author>Canu, Stéphane,</author>
		<affiliation>University of São Paulo</affiliation>
		<affiliation>University of São Paulo</affiliation>
		<affiliation>University of São Paulo</affiliation>
		<affiliation>LITIS, INSA de Rouen</affiliation>
		<editor>Aliaga, Daniel G.,</editor>
		<editor>Davis, Larry S.,</editor>
		<editor>Farias, Ricardo C.,</editor>
		<editor>Fernandes, Leandro A. F.,</editor>
		<editor>Gibson, Stuart J.,</editor>
		<editor>Giraldi, Gilson A.,</editor>
		<editor>Gois, João Paulo,</editor>
		<editor>Maciel, Anderson,</editor>
		<editor>Menotti, David,</editor>
		<editor>Miranda, Paulo A. V.,</editor>
		<editor>Musse, Soraia,</editor>
		<editor>Namikawa, Laercio,</editor>
		<editor>Pamplona, Mauricio,</editor>
		<editor>Papa, João Paulo,</editor>
		<editor>Santos, Jefersson dos,</editor>
		<editor>Schwartz, William Robson,</editor>
		<editor>Thomaz, Carlos E.,</editor>
		<e-mailaddress>igordsm@ime.usp.br</e-mailaddress>
		<conferencename>Conference on Graphics, Patterns and Images, 29 (SIBGRAPI)</conferencename>
		<conferencelocation>São José dos Campos, SP, Brazil</conferencelocation>
		<date>4-7 Oct. 2016</date>
		<publisher>IEEE Computer Society´s Conference Publishing Services</publisher>
		<publisheraddress>Los Alamitos</publisheraddress>
		<booktitle>Proceedings</booktitle>
		<tertiarytype>Full Paper</tertiarytype>
		<transferableflag>1</transferableflag>
		<versiontype>finaldraft</versiontype>
		<keywords>Kernel approximation, W-operator learning, Machine learning, Image Processing.</keywords>
		<abstract>Designing image operators is a hard task usually tackled by specialists in image processing. An alternative approach is to use machine learning to estimate local transformations, that characterize the image operators, from pairs of input-output images. The main challenge of this approach, called $W$-operator learning, is estimating operators over large windows without overfitting. Current techniques require the determination of a large number of parameters to maximize the performance of the trained operators. Support Vector Machines are known for their generalization performance and their ability to estimate nonlinear decision surfaces using kernels. However, training kernelized SVMs in the dual is not feasible when the training set is large. We estimate the local transformations employing kernel approximations to train SVMs, thus with no need to compute the full Gram matrix. We also select appropriate kernels to process binary and gray level inputs. Experiments show that operators trained using kernel approximation achieve comparable results with state-of-the-art methods in 4 public datasets.</abstract>
		<language>en</language>
		<targetfile>PID4373017.pdf</targetfile>
		<usergroup>igordsm@ime.usp.br</usergroup>
		<visibility>shown</visibility>
		<documentstage>not transferred</documentstage>
		<mirrorrepository>sid.inpe.br/banon/2001/03.30.15.38.24</mirrorrepository>
		<nexthigherunit>8JMKD3MGPAW/3M2D4LP</nexthigherunit>
		<nexthigherunit>8JMKD3MGPEW34M/4742MCS</nexthigherunit>
		<citingitemlist>sid.inpe.br/sibgrapi/2016/07.02.23.50 4</citingitemlist>
		<citingitemlist>sid.inpe.br/sibgrapi/2022/06.10.21.49 2</citingitemlist>
		<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/2016/07.22.14.26</url>
	</metadata>
</metadatalist>