Trust region versus line search for computing the optical flow

dc.contributor.authorKalmoun, El Mostafa
dc.contributor.authorGarrido Ostermann, Lluís
dc.date.accessioned2014-03-18T10:16:23Z
dc.date.available2014-03-18T10:16:23Z
dc.date.issued2013-09-12
dc.date.updated2014-03-18T10:16:23Z
dc.description.abstractWe consider the numerical treatment of the optical flow problem by evaluating the performance of the trust region method versus the line search method. To the best of our knowledge, the trust region method is studied here for the first time for variational optical flow computation. Four different optical flow models are used to test the performance of the proposed algorithm combining linear and nonlinear data terms with quadratic and TV regularization. We show that trust region often performs better than line search; especially in the presence of non-linearity and non-convexity in the model.
dc.format.extent17 p.
dc.format.mimetypeapplication/pdf
dc.identifier.idgrec635635
dc.identifier.issn1540-3459
dc.identifier.urihttps://hdl.handle.net/2445/52216
dc.language.isoeng
dc.publisherSociety for Industrial and Applied Mathematics
dc.relation.isformatofReproducció del document publicat a: http://dx.doi.org/10.1137/120897213
dc.relation.ispartofMultiscale Modeling & Simulation, 2013, vol. 11, num. 3, p. 890-906
dc.relation.urihttp://dx.doi.org/10.1137/120897213
dc.rights(c) Society for Industrial and Applied Mathematics., 2013
dc.rights.accessRightsinfo:eu-repo/semantics/openAccess
dc.sourceArticles publicats en revistes (Matemàtiques i Informàtica)
dc.subject.classificationInformàtica
dc.subject.classificationProgramació (Matemàtica)
dc.subject.classificationAnàlisi numèrica
dc.subject.otherComputer science
dc.subject.otherMathematical programming
dc.subject.otherNumerical analysis
dc.titleTrust region versus line search for computing the optical flow
dc.typeinfo:eu-repo/semantics/article
dc.typeinfo:eu-repo/semantics/publishedVersion

Fitxers

Paquet original

Mostrant 1 - 1 de 1
Carregant...
Miniatura
Nom:
635635.pdf
Mida:
272.96 KB
Format:
Adobe Portable Document Format