Show simple item record

dc.contributor.authorTartavel, Guillaume*
dc.contributor.authorGousseau, Yann*
dc.contributor.authorPeyré, Gabriel*
dc.date.accessioned2013-11-27T15:04:19Z
dc.date.available2013-11-27T15:04:19Z
dc.date.issued2015
dc.identifier.issn0924-9907
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/12192
dc.language.isoenen
dc.subjectExemplar-based synthesis
dc.subjectNon-convex optimization
dc.subjectStatistical image modeling
dc.subjectSparse decomposition
dc.subjectDictionary learning
dc.subjectRandom phase textures
dc.subjectGaussian random fields
dc.subject.ddc006.3en
dc.titleVariational Texture Synthesis with Sparsity and Spectrum Constraints
dc.typeArticle accepté pour publication ou publié
dc.contributor.editoruniversityotherTélécom ParisTech http://www.telecom-paristech.fr/ Institut Mines-Télécom;France
dc.contributor.editoruniversityotherLaboratoire Traitement et Communication de l'Information [Paris] (LTCI) http://www.ltci.telecom-paristech.fr/ Télécom ParisTech – CNRS : UMR5141;France
dc.description.abstractenThis paper introduces a new approach for texture synthesis. We propose a unified framework that both imposes first order statistical constraints on the use of atoms from an adaptive dictionary, as well as second order constraints on pixel values. This is achieved thanks to a variational approach, the minimization of which yields local extrema, each one being a possible texture synthesis. On the one hand, the adaptive dictionary is created using a sparse image representation rationale, and a global constraint is imposed on the maximal number of use of each atom from this dictionary. On the other hand, a constraint on second order pixel statistics is achieved through the power spectrum of images. In a sense, this method reconciles the first and second theory proposed by B. Julesz for texture discrimination. An advantage of the proposed method is its ability to truly synthesize textures, without verbatim copy of small pieces from the exemplar. In an extensive experimental section, we show that the resulting synthesis achieves state of the art results, both for structured and small scale textures.en
dc.relation.isversionofjnlnameJournal of Mathematical Imaging and Vision
dc.relation.isversionofjnlvol52
dc.relation.isversionofjnlissue1
dc.relation.isversionofjnldate2015
dc.relation.isversionofjnlpages124-144
dc.relation.isversionofdoihttp://dx.doi.org/10.1007/s10851-014-0547-7
dc.identifier.urlsitehttp://hal.archives-ouvertes.fr/hal-00881847en
dc.relation.isversionofjnlpublisherSpringeren
dc.subject.ddclabelIntelligence artificielleen
dc.relation.forthcomingprintouien
dc.description.ssrncandidatenon
dc.description.halcandidateoui
dc.description.readershiprecherche
dc.description.audienceinternational
dc.relation.Isversionofjnlpeerreviewedoui
dc.date.updated2016-01-04T15:18:31Z
hal.person.labIds*
hal.person.labIds*
hal.person.labIds*


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record