Skip navigation
Skip navigation

Efficient dense subspace clustering

Ji, Pan; Salzmann, Mathieu; Li, Hongdong

Description

In this paper, we tackle the problem of clustering data points drawn from a union of linear (or affine) subspaces. To this end, we introduce an efficient subspace clustering algorithm that estimates dense connections between the points lying in the same subspace. In particular, instead of following the standard compressive sensing approach, we formulate subspace clustering as a Frobenius norm minimization problem, which inherently yields denser con- nections between the data points. While in...[Show more]

dc.contributor.authorJi, Pan
dc.contributor.authorSalzmann, Mathieu
dc.contributor.authorLi, Hongdong
dc.coverage.spatialSteamboat Springs USA
dc.date.accessioned2015-12-13T22:29:16Z
dc.date.createdMarch 24-26 2014
dc.identifier.isbn9781479949854
dc.identifier.urihttp://hdl.handle.net/1885/74616
dc.description.abstractIn this paper, we tackle the problem of clustering data points drawn from a union of linear (or affine) subspaces. To this end, we introduce an efficient subspace clustering algorithm that estimates dense connections between the points lying in the same subspace. In particular, instead of following the standard compressive sensing approach, we formulate subspace clustering as a Frobenius norm minimization problem, which inherently yields denser con- nections between the data points. While in the noise-free case we rely on the self-expressiveness of the observations, in the presence of noise we simultaneously learn a clean dictionary to represent the data. Our formulation lets us address the subspace clustering problem efficiently. More specifically, the solution can be obtained in closed-form for outlier-free observations, and by performing a series of linear operations in the presence of outliers. Interestingly, we show that our Frobenius norm formulation shares the same solution as the popular nuclear norm minimization approach when the data is free of any noise, or, in the case of corrupted data, when a clean dictionary is learned. Our experimental evaluation on motion segmentation and face clustering demonstrates the benefits of our algorithm in terms of clustering accuracy and efficiency.
dc.publisherIEEE Computer Society
dc.relation.ispartofseries2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
dc.source2014 IEEE Winter Conference on Applications of Computer Vision, WACV 2014
dc.titleEfficient dense subspace clustering
dc.typeConference paper
local.description.notesImported from ARIES
local.description.refereedYes
dc.date.issued2014
local.identifier.absfor080104 - Computer Vision
local.identifier.ariespublicationU3488905xPUB4209
local.type.statusPublished Version
local.contributor.affiliationJi, Pan, College of Engineering and Computer Science, ANU
local.contributor.affiliationSalzmann, Mathieu, College of Engineering and Computer Science, ANU
local.contributor.affiliationLi, Hongdong, College of Engineering and Computer Science, ANU
local.description.embargo2037-12-31
local.bibliographicCitation.startpage461
local.bibliographicCitation.lastpage468
local.identifier.doi10.1109/WACV.2014.6836065
dc.date.updated2015-12-11T08:47:43Z
local.identifier.scopusID2-s2.0-84904615403
CollectionsANU Research Publications

Download

File Description SizeFormat Image
01_Ji_Efficient_dense_subspace_2014.pdf886.66 kBAdobe PDF    Request a copy


Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.

Updated:  22 January 2019/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator