Multi-view Low-rank Sparse Subspace Clustering

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 importance 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 encourages 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 problem 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.

References in corpus (4)

• Robust Recovery of Subspace Structures by Low-Rank Representation
• A Survey on Multi-view Learning
• Convex Sparse Spectral Clustering: Single-view to Multi-view
• Multi-View Matrix Completion for Multi-Label Image Classification

Cited by in corpus (25)

• Essential Tensor Learning for Multi-view Spectral Clustering
• Deep Embedded Multi-view Clustering with Collaborative Training
• Feature Concatenation Multi-view Subspace Clustering
• ${\ell }_0$-Motivated Low-Rank Sparse Subspace Clustering
• Unbalanced Incomplete Multi-view Clustering via the Scheme of View Evolution: Weak Views are Meat; Strong Views do Eat
• A Nonlinear Orthogonal Non-Negative Matrix Factorization Approach to Subspace Clustering
• Latent Complete Row Space Recovery for Multi-view Subspace Clustering
• Enhanced Latent Multi-view Subspace Clustering
• Constrained Bilinear Factorization Multi-view Subspace Clustering
• Large-scale Multi-view Subspace Clustering in Linear Time
• Efficient Parameter-free Clustering Using First Neighbor Relations
• Disentangling Multi-view Representations Beyond Inductive Bias
• Structured Graph Learning for Scalable Subspace Clustering: From Single-view to Multi-view
• Unsupervised Multi-view Clustering by Squeezing Hybrid Knowledge from Cross View and Each View
• Multi-view Contrastive Graph Clustering
• Multi-view Subspace Clustering Networks with Local and Global Graph Information
• Robust Self-Supervised Convolutional Neural Network for Subspace Clustering and Classification
• Tensor-based Intrinsic Subspace Representation Learning for Multi-view Clustering
• Smoothed Multi-View Subspace Clustering
• Joint Featurewise Weighting and Lobal Structure Learning for Multi-view Subspace Clustering
• Consistent and Complementary Graph Regularized Multi-view Subspace Clustering
• Adaptive Weighted Multi-View Clustering
• Multi-view Subspace Adaptive Learning via Autoencoder and Attention
• Kernel Two-Dimensional Ridge Regression for Subspace Clustering
• Discriminative community detection for multiplex networks