Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/140530
Full metadata record
DC FieldValueLanguage
dc.contributor.authorCai, Yujunen_US
dc.contributor.authorGe, Liuhaoen_US
dc.contributor.authorCai, Jianfeien_US
dc.contributor.authorYuan, Junsongen_US
dc.date.accessioned2020-05-30T07:09:34Z-
dc.date.available2020-05-30T07:09:34Z-
dc.date.issued2018-
dc.identifier.citationCai, Y., Ge, L., Cai, J., & Yuan, J. (2018). Weakly-supervised 3D hand pose estimation from monocular RGB images. European Conference on Computer Vision (ECCV) 2018, 678-694. doi:10.1007/978-3-030-01231-1_41en_US
dc.identifier.urihttps://hdl.handle.net/10356/140530-
dc.description.abstractCompared with depth-based 3D hand pose estimation, it is more challenging to infer 3D hand pose from monocular RGB images, due to substantial depth ambiguity and the difficulty of obtaining fully-annotated training data. Different from existing learning-based monocular RGB-input approaches that require accurate 3D annotations for training, we propose to leverage the depth images that can be easily obtained from commodity RGB-D cameras during training, while during testing we take only RGB inputs for 3D joint predictions. In this way, we alleviate the burden of the costly 3D annotations in real-world dataset. Particularly, we propose a weakly-supervised method, adaptating from fully-annotated synthetic dataset to weakly-labeled real-world dataset with the aid of a depth regularizer, which generates depth maps from predicted 3D pose and serves as weak supervision for 3D pose regression. Extensive experiments on benchmark datasets validate the effectiveness of the proposed depth regularizer in both weakly-supervised and fully-supervised settings.en_US
dc.description.sponsorshipNRF (Natl Research Foundation, S’pore)en_US
dc.description.sponsorshipMOE (Min. of Education, S’pore)en_US
dc.language.isoenen_US
dc.relationMOE2016-T2-2-065en_US
dc.rights© 2018 Springer Nature Switzerland AG. All rights reserved. This paper was published in European Conference on Computer Vision (ECCV) 2018 and is made available with permission of Springer Nature Switzerland AG.en_US
dc.subjectEngineering::Computer science and engineeringen_US
dc.titleWeakly-supervised 3D hand pose estimation from monocular RGB imagesen_US
dc.typeConference Paperen
dc.contributor.schoolSchool of Computer Science and Engineeringen_US
dc.contributor.schoolInterdisciplinary Graduate School (IGS)en_US
dc.contributor.conferenceEuropean Conference on Computer Vision (ECCV) 2018en_US
dc.contributor.organizationDepartment of Computer Science and Engineering, State University of New York at Buffalo Universityen_US
dc.contributor.researchInstitute for Media Innovation (IMI)en_US
dc.identifier.doi10.1007/978-3-030-01231-1_41-
dc.description.versionAccepted versionen_US
dc.identifier.spage678en_US
dc.identifier.epage694en_US
dc.subject.keywordsComputer Visionen_US
dc.subject.keywords3D Hand Pose Estimationen_US
item.grantfulltextopen-
item.fulltextWith Fulltext-
Appears in Collections:IMI Conference Papers
Files in This Item:
File Description SizeFormat 
Weakly-supervised 3D Hand Pose Estimation from Monocular RGB Images.pdf2.39 MBAdobe PDFThumbnail
View/Open

SCOPUSTM   
Citations 20

5
Updated on Jul 13, 2020

PublonsTM
Citations 20

7
Updated on Mar 9, 2021

Page view(s)

69
Updated on May 14, 2021

Download(s)

3
Updated on May 14, 2021

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.