Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorLu, Xuequanen_US
dc.contributor.authorDeng, Zhigangen_US
dc.contributor.authorLuo, Junen_US
dc.contributor.authorChen, Wenzhien_US
dc.contributor.authorYeung, Sai-Kiten_US
dc.contributor.authorHe, Yingen_US
dc.identifier.citationLu, X., Deng, Z., Luo, J., Chen, W., Yeung, S. & He, Y. (2019). 3D articulated skeleton extraction using a single consumer-grade depth camera. Computer Vision and Image Understanding, 188, 102792-.
dc.description.abstractArticulated skeleton extraction or learning has been extensively studied for 2D (e.g., images and video) and 3D (e.g., volume sequences, motion capture, and mesh sequences) data. Nevertheless, robustly and accurately learning 3D articulated skeletons from point set sequences captured by a single consumer-grade depth camera still remains challenging, since such data are often corrupted with substantial noise and outliers. Relatively few approaches have been proposed to tackle this problem. In this paper, we present a novel unsupervised framework to address this issue. Specifically, we first build one-to-one point correspondences among the point cloud frames in a sequence with our non-rigid point cloud registration algorithm. We then generate a skeleton involving a reasonable number of joints and bones with our skeletal structure extraction algorithm. We lastly present an iterative Linear Blend Skinning based algorithm for accurate joints learning. At the end, our method can learn a quality articulated skeleton from a single 3D point sequence possibly corrupted with noise and outliers. Through qualitative and quantitative evaluations on both publicly available data and in-house Kinect-captured data, we show that our unsupervised approach soundly outperforms state of the art techniques in terms of both quality (i.e., visual) and accuracy (i.e., Euclidean distance error metric). Moreover, the poses of our extracted skeletons are even comparable to those by KinectSDK, a well-known supervised pose estimation technique; for example, our method and KinectSDK achieves similar distance errors of 0.0497 and 0.0521.en_US
dc.description.sponsorshipMinistry of Education (MOE)en_US
dc.relationMOE RG26/17en_US
dc.relation.ispartofComputer Vision and Image Understandingen_US
dc.rights© 2019 Elsevier Inc. All rights reserved.en_US
dc.subjectEngineering::Computer science and engineeringen_US
dc.title3D articulated skeleton extraction using a single consumer-grade depth cameraen_US
dc.typeJournal Articleen
dc.contributor.schoolSchool of Computer Science and Engineeringen_US
dc.subject.keywordsUnsupervised Skeleton Extractionen_US
dc.subject.keywordsSingle Viewen_US
dc.description.acknowledgementXuequan Lu is supported in part by Deakin University, Australia CY01-251301-F003-PJ03906-PG00447. Zhigang Deng is in part supported by NSF, USA IIS-1524782. Jun Luo is supported in part by AcRF Tier 2 Grant MOE2016-T2-2-022 (Singapore). Ying He is supported by MOE RG26/17. Sai-Kit Yeung is supported by an internal grant from HKUST (R9429).en_US
item.fulltextNo Fulltext-
Appears in Collections:SCSE Journal Articles

Citations 20

Updated on Jan 29, 2023

Web of ScienceTM
Citations 20

Updated on Jan 30, 2023

Page view(s)

Updated on Feb 2, 2023

Google ScholarTM




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