Full metadata record
DC FieldValueLanguage
dc.contributor.authorLin, Jia-Ruen_US
dc.contributor.authorLin, I-Chenen_US
dc.date.accessioned2014-12-08T15:38:08Z-
dc.date.available2014-12-08T15:38:08Z-
dc.date.issued2011-01-01en_US
dc.identifier.issn1016-2364en_US
dc.identifier.urihttp://hdl.handle.net/11536/26157-
dc.description.abstractIn this paper, a feature-point-driven expression editing and synthesis framework is proposed. While the extensively used blend shape methods suffer detail losing during image blending, the proposed multi-layer method can retain the flexibility and variety of geometry editing and preserve detail features as well. For low-frequency sub-bands, optimization-based blend shape is presented for large-to-mid scale synthesis. In addition, statistics-based feature matching and enhancement are proposed for high-frequency details. Our results show that the proposed methods are adequate to high-resolution expression synthesis and detail-preserved image editing.en_US
dc.language.isoen_USen_US
dc.subjectfacial expressionen_US
dc.subjectblend shapeen_US
dc.subjectimage editingen_US
dc.subjectsteerable pyramiden_US
dc.subjectdetail preservationen_US
dc.titleMulti-layered Expression Synthesisen_US
dc.typeArticleen_US
dc.identifier.journalJOURNAL OF INFORMATION SCIENCE AND ENGINEERINGen_US
dc.citation.volume27en_US
dc.citation.issue1en_US
dc.citation.spage337en_US
dc.citation.epage351en_US
dc.contributor.department多媒體工程研究所zh_TW
dc.contributor.departmentInstitute of Multimedia Engineeringen_US
dc.identifier.wosnumberWOS:000286692500022-
dc.citation.woscount0-
Appears in Collections:Articles


Files in This Item:

  1. 000286692500022.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.