Multi-view low-rank sparse subspace clustering

Brbić, Maria; Kopriva, Ivica (2018) Multi-view low-rank sparse subspace clustering. Pattern Recognition, 73 (1). pp. 247-258. ISSN 0031-3203

Preview

PDF - Accepted Version - article Download (1MB) | Preview

Abstract

Most existing approaches address multi-view subspace clustering problem by constructing the affinity matrix on each view separately and afterwards propose how to extend spectral clustering algorithm to handle multi-view data. This paper presents an approach to multi-view subspace clustering that learns a joint subspace representation by constructing affinity matrix shared among all views. Relying on the im- portance of both low-rank and sparsity constraints in the construction of the affinity matrix, we introduce the objective that balances between the agreement across different views, while at the same time encour- ages sparsity and low-rankness of the solution. Related low-rank and sparsity constrained optimization problem is for each view solved using the alternating direction method of multipliers. Furthermore, we extend our approach to cluster data drawn from nonlinear subspaces by solving the corresponding prob- lem in a reproducing kernel Hilbert space. The proposed algorithm outperforms state-of-the-art multi- view subspace clustering algorithms on one synthetic and four real-world datasets.

Item Type:	Article
Additional Information:	This work has been supported by the Croatian Science Foundation grant IP-2016-06-5235 (Structured Decompositions of Empirical Data for Computationally-Assisted Diagnoses of Disease) and by the Croatian Science Foundation grant HRZZ-9623 (Descriptive Induction).
Uncontrolled Keywords:	Subspace clustering; Multi-view data; Low-rank; Sparsity; Alternating direction method of multipliers; Reproducing kernel Hilbert space
Subjects:	NATURAL SCIENCES > Mathematics > Applied Mathematics and Mathematical Modeling TECHNICAL SCIENCES
Divisions:	Division of Electronics
Projects:	Project title Project leader Project code Project type Strukturne dekompozicije empirijskih podataka za računalno potpomognutu dijagnostiku bolesti-DEDAD Ivica Kopriva IP-2016-06-5235 HRZZ
Depositing User:	Ivica Kopriva
Date Deposited:	26 Nov 2019 09:03
URI:	http://fulir.irb.hr/id/eprint/5147
DOI:	10.1016/j.patcog.2017.08.024

Actions (login required)

View Item