Close

1. Identity statement
Reference TypeConference Paper (Conference Proceedings)
Sitesibgrapi.sid.inpe.br
Holder Codeibi 8JMKD3MGPEW34M/46T9EHH
Identifier6qtX3pFwXQZeBBx/vRS9M
Repositorysid.inpe.br/banon/2002/10.24.09.47
Last Update2002:09.23.03.00.00 (UTC) administrator
Metadata Repositorysid.inpe.br/banon/2002/10.24.09.47.55
Metadata Last Update2022:06.14.00.11.49 (UTC) administrator
DOI10.1109/SIBGRA.2002.1167155
Citation KeyCaetanoOlabBaro:2002:PeEvSi
TitlePerformance evaluation of single and multiple-gaussian models for skin color modeling
Year2002
Access Date2024, May 02
Number of Files1
Size1078 KiB
2. Context
Author1 Caetano, Tibério S.
2 Olabarriaga, Sílvia D.
3 Barone, Dante A. C.
EditorGonçalves, Luiz Marcos Garcia
Musse, Soraia Raupp
Comba, João Luiz Dihl
Giraldi, Gilson
Dreux, Marcelo
Conference NameBrazilian Symposium on Computer Graphics and Image Processing, 15 (SIBGRAPI)
Conference LocationFortaleza, CE, Brazil
Date10-10 Oct. 2002
PublisherIEEE Computer Society
Publisher CityLos Alamitos
Book TitleProceedings
Tertiary TypeFull Paper
OrganizationSBC - Brazilian Computer Society
History (UTC)2008-07-17 14:10:46 :: administrator -> banon ::
2008-08-26 15:21:22 :: banon -> administrator ::
2009-08-13 20:36:39 :: administrator -> banon ::
2010-08-28 20:00:06 :: banon -> administrator ::
2022-06-14 00:11:49 :: administrator -> :: 2002
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
Version Typefinaldraft
AbstractWe present an experimental setup to evaluate the relative performance of single gaussian and mixture of gaussians models for skin color modeling. Firstly, a sample set of 1,120,000 skin pixels from a number of ethnic groups is selected and represented in the chromaticity space. In the following, parameter estimation for both the single gaussian and seven (with 2 to 8 gaussian components) gaussian mixture models is performed. For the mixture models, learning is carried out via the expectation-maximisation (EM) algorithm. In order to compare performances achieved by the 8 different models, we apply to each model a test set of 800 images-none from the training set. True skin regions, representing the ground truth, are manually selected, and false positive and true positive rates are computed for each value of a specific threshold. Finally, receiver operating characteristics (ROC) curves are plotted for each model, which make it possible to analyze and compare their relative performances. Results obtained show that, for medium to high true positive rates, mixture models (with 2 to 8 components) outperform the single gaussian model. Nevertheless, for low false positive rates, all the models behave similarly.
Arrangement 1urlib.net > SDLA > Fonds > SIBGRAPI 2002 > Performance evaluation of...
Arrangement 2urlib.net > SDLA > Fonds > Full Index > Performance evaluation of...
doc Directory Contentaccess
source Directory Contentthere are no files
agreement Directory Contentthere are no files
4. Conditions of access and use
data URLhttp://urlib.net/ibi/6qtX3pFwXQZeBBx/vRS9M
zipped data URLhttp://urlib.net/zip/6qtX3pFwXQZeBBx/vRS9M
Languageen
Target File69.pdf
User Groupadministrator
Visibilityshown
5. Allied materials
Next Higher Units8JMKD3MGPEW34M/46QCSHP
8JMKD3MGPEW34M/4742MCS
Citing Item Listsid.inpe.br/sibgrapi/2022/05.01.04.11 7
Host Collectionsid.inpe.br/banon/2001/03.30.15.38
6. Notes
NotesThe conference was held in Fortaleza, CE, Brazil, from October 7 to 10.
Empty Fieldsaffiliation archivingpolicy archivist area callnumber contenttype copyholder copyright creatorhistory descriptionlevel dissemination documentstage e-mailaddress edition electronicmailaddress format group isbn issn keywords label lineage mark mirrorrepository nextedition numberofvolumes orcid 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