3-D motion recovery via low rank matrix analysis

Wang, Meiyuan, Li, Kun, Wu, Feng, Lai, Yukun ORCID: https://orcid.org/0000-0002-2094-5680 and Jingyu, Yang 2017. 3-D motion recovery via low rank matrix analysis. Presented at: IEEE International Conference on Visual Communications and Image Processing (VCIP), Chengdu, China, 27-30 November 2016. 2016 Visual Communications and Image Processing (VCIP). IEEE, pp. 1-4. 10.1109/VCIP.2016.7805473

Preview

PDF - Submitted Pre-Print Version Download (1MB) | Preview

Abstract

Skeleton tracking is a useful and popular application of Kinect. However, it cannot provide accurate reconstructions for complex motions, especially in the presence of occlusion. This paper proposes a new 3-D motion recovery method based on lowrank matrix analysis to correct invalid or corrupted motions. We address this problem by representing a motion sequence as a matrix, and introducing a convex low-rank matrix recovery model, which fixes erroneous entries and finds the correct low-rank matrix by minimizing nuclear norm and `1-norm of constituent clean motion and error matrices. Experimental results show that our method recovers the corrupted skeleton joints, achieving accurate and smooth reconstructions even for complicated motions.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Publication
Status:	Published
Schools:	Computer Science & Informatics
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Publisher:	IEEE
ISBN:	9781509053179
Related URLs:	Organisation
Date of First Compliant Deposit:	27 August 2016
Date of Acceptance:	7 August 2016
Last Modified:	01 Nov 2022 11:10
URI:	https://orca.cardiff.ac.uk/id/eprint/94061

Citation Data

Cited 8 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item