DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Seong-heum | - |
dc.contributor.author | Tai, Yu-Wing | - |
dc.contributor.author | Lee, Joon-Young | - |
dc.contributor.author | Park, Jaesik | - |
dc.contributor.author | Kweon, In So | - |
dc.date.accessioned | 2019-12-23T09:10:03Z | - |
dc.date.available | 2019-12-23T09:10:03Z | - |
dc.date.created | 2019-12-04 | - |
dc.date.issued | 2017-12 | - |
dc.identifier.issn | 0167-7055 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/100558 | - |
dc.description.abstract | In this paper, we present a new framework to determine up front orientations and detect salient views of 3D models. The salient viewpoint to human preferences is the most informative projection with correct upright orientation. Our method utilizes two Convolutional Neural Network (CNN) architectures to encode category-specific information learnt from a large number of 3D shapes and 2D images on the web. Using the first CNN model with 3D voxel data, we generate a CNN shape feature to decide natural upright orientation of 3D objects. Once a 3D model is upright-aligned, the front projection and salient views are scored by category recognition using the second CNN model. The second CNN is trained over popular photo collections from internet users. In order to model comfortable viewing angles of 3D models, a category-dependent prior is also learnt from the users. Our approach effectively combines category-specific scores and classical evaluations to produce a data-driven viewpoint saliency map. The best viewpoints from the method are quantitatively and qualitatively validated with more than 100 objects from 20 categories. Our thumbnail images of 3D models are the most favoured among those from different approaches. | - |
dc.language | English | - |
dc.publisher | Blackwell Publishing Inc. | - |
dc.relation.isPartOf | Computer Graphics Forum | - |
dc.title | Category‐Specific Salient View Selection via Deep Convolutional Neural Networks | - |
dc.type | Article | - |
dc.identifier.doi | 10.1111/cgf.13082 | - |
dc.type.rims | ART | - |
dc.identifier.bibliographicCitation | Computer Graphics Forum, v.36, no.8, pp.313 - 328 | - |
dc.identifier.wosid | 000417496200022 | - |
dc.citation.endPage | 328 | - |
dc.citation.number | 8 | - |
dc.citation.startPage | 313 | - |
dc.citation.title | Computer Graphics Forum | - |
dc.citation.volume | 36 | - |
dc.contributor.affiliatedAuthor | Park, Jaesik | - |
dc.identifier.scopusid | 2-s2.0-85013249519 | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | N | - |
dc.type.docType | Article | - |
dc.subject.keywordPlus | MENTAL ROTATION | - |
dc.subject.keywordPlus | 3D | - |
dc.subject.keywordPlus | ORIENTATION | - |
dc.subject.keywordPlus | IMAGE | - |
dc.subject.keywordAuthor | best view selection | - |
dc.subject.keywordAuthor | upright orientation estimation | - |
dc.subject.keywordAuthor | deep learning | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Software Engineering | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.