Close
Metadata

1. Identity statement
Reference TypeConference Paper (Conference Proceedings)
Sitesibgrapi.sid.inpe.br
Holder Codeibi 8JMKD3MGPEW34M/46T9EHH
Identifier8JMKD3MGPAW/3M5J6KE
Repositorysid.inpe.br/sibgrapi/2016/07.22.14.26
Last Update2016:07.22.14.37.43 (UTC) administrator
Metadata Repositorysid.inpe.br/sibgrapi/2016/07.22.14.26.25
Metadata Last Update2022:06.14.00.08.34 (UTC) administrator
DOI10.1109/SIBGRAPI.2016.060
Citation KeyMontagnerJrHiraCanu:2016:KeApWo
TitleKernel approximations for W-operator learning
FormatOn-line
Year2016
Access Date2024, May 02
Number of Files1
Size753 KiB
2. Context
Author1 Montagner, Igor S.
2 Jr., Roberto Hirata
3 Hirata, Nina S. T.
4 Canu, Stéphane
Affiliation1 University of São Paulo
2 University of São Paulo
3 University of São Paulo
4 LITIS, INSA de Rouen
EditorAliaga, Daniel G.
Davis, Larry S.
Farias, Ricardo C.
Fernandes, Leandro A. F.
Gibson, Stuart J.
Giraldi, Gilson A.
Gois, João Paulo
Maciel, Anderson
Menotti, David
Miranda, Paulo A. V.
Musse, Soraia
Namikawa, Laercio
Pamplona, Mauricio
Papa, João Paulo
Santos, Jefersson dos
Schwartz, William Robson
Thomaz, Carlos E.
e-Mail Addressigordsm@ime.usp.br
Conference NameConference on Graphics, Patterns and Images, 29 (SIBGRAPI)
Conference LocationSão José dos Campos, SP, Brazil
Date4-7 Oct. 2016
PublisherIEEE Computer Society´s Conference Publishing Services
Publisher CityLos Alamitos
Book TitleProceedings
Tertiary TypeFull Paper
History (UTC)2016-07-22 14:37:43 :: igordsm@ime.usp.br -> administrator :: 2016
2016-10-05 14:49:16 :: administrator -> igordsm@ime.usp.br :: 2016
2016-10-21 13:35:26 :: igordsm@ime.usp.br -> administrator :: 2016
2022-06-14 00:08:34 :: administrator -> :: 2016
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
Version Typefinaldraft
KeywordsKernel approximation
W-operator learning
Machine learning
Image Processing
AbstractDesigning 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.
Arrangement 1urlib.net > SDLA > Fonds > SIBGRAPI 2016 > Kernel approximations for...
Arrangement 2urlib.net > SDLA > Fonds > Full Index > Kernel approximations for...
doc Directory Contentaccess
source Directory Content
PID4373017.pdf 22/07/2016 11:26 752.5 KiB 
agreement Directory Content
agreement.html 22/07/2016 11:26 1.2 KiB 
4. Conditions of access and use
data URLhttp://urlib.net/ibi/8JMKD3MGPAW/3M5J6KE
zipped data URLhttp://urlib.net/zip/8JMKD3MGPAW/3M5J6KE
Languageen
Target FilePID4373017.pdf
User Groupigordsm@ime.usp.br
Visibilityshown
Update Permissionnot transferred
5. Allied materials
Mirror Repositorysid.inpe.br/banon/2001/03.30.15.38.24
Next Higher Units8JMKD3MGPAW/3M2D4LP
8JMKD3MGPEW34M/4742MCS
Citing Item Listsid.inpe.br/sibgrapi/2016/07.02.23.50 4
sid.inpe.br/sibgrapi/2022/06.10.21.49 2
Host Collectionsid.inpe.br/banon/2001/03.30.15.38
6. Notes
Empty Fieldsarchivingpolicy archivist area callnumber contenttype copyholder copyright creatorhistory descriptionlevel dissemination edition electronicmailaddress group isbn issn label lineage mark nextedition notes numberofvolumes orcid organization pages parameterlist parentrepositories previousedition previouslowerunit progress project readergroup readpermission resumeid rightsholder schedulinginformation secondarydate secondarykey secondarymark secondarytype serieseditor session shorttitle sponsor subject tertiarymark type url volume


Close