Fine detailed texture learning for 3D meshes with generative models

Simple item page
buir.contributor.authorDündar , Ayşegül
buir.contributor.orcidDündar, Ayşegül|0000-0003-2014-6325
dc.citation.epage14574en_US
dc.citation.issueNumber12
dc.citation.spage14563
dc.citation.volumeNumber45
dc.contributor.authorDündar, Ayşegül
dc.contributor.authorGao, J.
dc.contributor.authorTao, A.
dc.contributor.authorCatanzaro, B.
dc.date.accessioned2024-03-19T10:31:21Z
dc.date.available2024-03-19T10:31:21Z
dc.date.issued2023-11-03
dc.departmentDepartment of Computer Engineering
dc.description.abstractThis paper presents a method to achieve fine detailed texture learning for 3D models that are reconstructed from both multi-view and single-view images. The framework is posed as an adaptation problem and is done progressively where in the first stage, we focus on learning accurate geometry, whereas in the second stage, we focus on learning the texture with a generative adversarial network. The contributions of the paper are in the generative learning pipeline where we propose two improvements. First, since the learned textures should be spatially aligned, we propose an attention mechanism that relies on the learnable positions of pixels. Second, since discriminator receives aligned texture maps, we augment its input with a learnable embedding which improves the feedback to the generator. We achieve significant improvements on multi-view sequences from Tripod dataset as well as on single-view image datasets, Pascal 3D+ and CUB. We demonstrate that our method achieves superior 3D textured models compared to the previous works.
dc.description.provenanceMade available in DSpace on 2024-03-19T10:31:21Z (GMT). No. of bitstreams: 1 Fine_detailed_texture_learning_for_3D_meshes_with_generative_models.pdf: 9206585 bytes, checksum: 15cac2ca4e6919bc720f6f67ceba07f3 (MD5) Previous issue date: 2023-12-01en
dc.identifier.doi10.1109/TPAMI.2023.3319429
dc.identifier.eissn1939-3539
dc.identifier.issn0162-8828
dc.identifier.urihttps://hdl.handle.net/11693/114968
dc.language.isoEnglish
dc.publisherInstitute of Electrical and Electronics Engineers
dc.relation.isversionofhttps://dx.doi.org/10.1109/TPAMI.2023.3319429
dc.source.titleIEEE Transactions on Pattern Analysis and Machine Intelligence
dc.subject3D texture learning
dc.subjectGenerative adversarial networks
dc.subject3D reconstruction
dc.titleFine detailed texture learning for 3D meshes with generative models
dc.typeArticle

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Fine_detailed_texture_learning_for_3D_meshes_with_generative_models.pdf
Size:
8.77 MB
Format:
Adobe Portable Document Format
Download

Collections

Scholarly Publications - Computer Engineering