Please use this identifier to cite or link to this item: https://repository.cihe.edu.hk/jspui/handle/cihe/4448
DC FieldValueLanguage
dc.contributor.authorLiu, Huien_US
dc.contributor.otherGuo, M.-
dc.contributor.otherHou, J.-
dc.contributor.otherJin, J.-
dc.contributor.otherZeng, H.-
dc.contributor.otherLu, J.-
dc.date.accessioned2024-04-10T06:06:54Z-
dc.date.available2024-04-10T06:06:54Z-
dc.date.issued2023-
dc.identifier.urihttps://repository.cihe.edu.hk/jspui/handle/cihe/4448-
dc.description.abstractExisting image-based rendering methods usually adopt depth-based image warping operation to synthesize novel views. In this paper, we reason the essential limitations of the traditional warping operation to be the limited neighborhood and only distance-based interpolation weights. To this end, we propose content-aware warping , which adaptively learns the interpolation weights for pixels of a relatively large neighborhood from their contextual information via a lightweight neural network. Based on this learnable warping module, we propose a new end-to-end learning-based framework for novel view synthesis from a set of input source views, in which two additional modules, namely confidence-based blending and feature-assistant spatial refinement, are naturally proposed to handle the occlusion issue and capture the spatial correlation among pixels of the synthesized view, respectively. Besides, we also propose a weight-smoothness loss term to regularize the network. Experimental results on light field datasets with wide baselines and multi-view datasets show that the proposed method significantly outperforms state-of-the-art methods both quantitatively and visually.en_US
dc.language.isoenen_US
dc.publisherIEEEen_US
dc.relation.ispartofIEEE Transactions on Pattern Analysis and Machine Intelligenceen_US
dc.titleContent-aware warping for view synthesisen_US
dc.typejournal articleen_US
dc.identifier.doi10.1109/TPAMI.2023.3242709-
dc.contributor.affiliationSchool of Computing and Information Sciencesen_US
dc.relation.issn1939-3539en_US
dc.description.volume45en_US
dc.description.issue8en_US
dc.description.startpage9486en_US
dc.description.endpage9503en_US
dc.cihe.affiliatedYes-
item.languageiso639-1en-
item.fulltextNo Fulltext-
item.openairecristypehttp://purl.org/coar/resource_type/c_6501-
item.cerifentitytypePublications-
item.openairetypejournal article-
item.grantfulltextnone-
crisitem.author.deptYam Pak Charitable Foundation School of Computing and Information Sciences-
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.