Please use this identifier to cite or link to this item: https://repository.cihe.edu.hk/jspui/handle/cihe/1253
DC FieldValueLanguage
dc.contributor.authorSiu, Wan Chien_US
dc.contributor.authorLiu, Zhisong-
dc.contributor.otherLi, C.-T.-
dc.contributor.otherWang, L.-W.-
dc.contributor.otherLun, D. P.-K.-
dc.date.accessioned2021-08-11T10:04:16Z-
dc.date.available2021-08-11T10:04:16Z-
dc.date.issued2020-
dc.identifier.urihttps://repository.cihe.edu.hk/jspui/handle/cihe/1253-
dc.description.abstractThe degree of difficulty in image inpainting depends on the types and sizes of the missing parts. Existing image inpainting approaches usually encounter difficulties in completing the missing parts in the wild with pleasing visual and contextual results as they are trained for either dealing with one specific type of missing patterns (mask) or unilaterally assuming the shapes and/or sizes of the masked areas. We propose a deep generative inpainting network, named DeepGIN, to handle various types of masked images. We design a Spatial Pyramid Dilation (SPD) ResNet block to enable the use of distant features for reconstruction. We also employ Multi-Scale Self-Attention (MSSA) mechanism and Back Projection (BP) technique to enhance our inpainting results. Our DeepGIN outperforms the state-of-the-art approaches generally, including two publicly available datasets (FFHQ and Oxford Buildings), both quantitatively and qualitatively. We also demonstrate that our model is capable of completing masked images in the wild.en_US
dc.language.isoenen_US
dc.publisherSpringeren_US
dc.titleDeepGIN: Deep generative inpainting network for extreme image inpaintingen_US
dc.typeconference proceedingsen_US
dc.relation.publicationComputer Vision – ECCV 2020 Workshops Proceedings, Part IVen_US
dc.identifier.doi10.1007/978-3-030-66823-5_1-
dc.contributor.affiliationSchool of Computing and Information Sciencesen_US
dc.relation.isbn9783030668228en_US
dc.description.startpage5en_US
dc.description.endpage22en_US
dc.cihe.affiliatedNo-
item.openairecristypehttp://purl.org/coar/resource_type/c_5794-
item.cerifentitytypePublications-
item.openairetypeconference proceedings-
item.fulltextNo Fulltext-
item.grantfulltextnone-
item.languageiso639-1en-
crisitem.author.deptSchool of Computing and Information Sciences-
crisitem.author.deptSchool of Computing and Information Sciences-
crisitem.author.orcid0000-0001-8280-0367-
crisitem.author.orcid0000-0003-4507-3097-
Appears in Collections:CIS Publication
SFX Query Show simple item record

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.