Improved subspace clustering via exploitation of spatial constraints

Improved subspace clustering via exploitation of spatial constraints

Pham, Duc-Son, Budhaditya, Saha, Phung, Dinh and Venkatesh, Svetha 2012, Improved subspace clustering via exploitation of spatial constraints, in CVPR 2012 : Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE, Piscataway, N. J., pp. 550-557.

Attached Files (Some files may be inaccessible until you login with your DRO credentials)
Name			Description	MIMEType		Size	Downloads

Title	Improved subspace clustering via exploitation of spatial constraints
Author(s)	Pham, Duc-Son Budhaditya, Saha Phung, Dinh Venkatesh, Svetha
Conference name	Computer Vision and Pattern Recognition. Conference (2012 : Providence, R. I.)
Conference location	Providence, R. I.
Conference dates	16-21 Jun. 2012
Title of proceedings	CVPR 2012 : Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Editor(s)	[Unknown]
Publication date	2012
Conference series	Computer Vision and Pattern Recognition. Conference (2012 : Providence, R. I.)
Start page	550
End page	557
Total pages	8
Publisher	IEEE
Place of publication	Piscataway, N. J.
Keyword(s)	computational costs corrupted data data sets Hopkins motion segmentation numerical algorithms sparse representation sparse solutions spatial constraints spatial geometry subspace clustering
Summary	We present a novel approach to improving subspace clustering by exploiting the spatial constraints. The new method encourages the sparse solution to be consistent with the spatial geometry of the tracked points, by embedding weights into the sparse formulation. By doing so, we are able to correct sparse representations in a principled manner without introducing much additional computational cost. We discuss alternative ways to treat the missing and corrupted data using the latest theory in robust lasso regression and suggest numerical algorithms so solve the proposed formulation. The experiments on the benchmark Johns Hopkins 155 dataset demonstrate that exploiting spatial constraints significantly improves motion segmentation.
ISBN	9781467312264 9781467312271
ISSN	1063-6919
Language	eng
Field of Research	080104 Computer Vision
Socio Economic Objective	970108 Expanding Knowledge in the Information and Computing Sciences
HERDC Research category	E1 Full written paper - refereed
Persistent URL	http://hdl.handle.net/10536/DRO/DU:30049552

Document type:	Conference Paper
Collection:	Centre for Pattern Recognition and Data Analytics

Citation counts:	Cited 1 times in Scopus Search Google Scholar
Access Statistics:	73 Abstract Views, 4 File Downloads - Detailed Statistics
Created:	Thu, 29 Nov 2012, 07:58:12 EST

Deakin Research Online

Improved subspace clustering via exploitation of spatial constraints

Study

Research

Access Deakin

On campus

Connect