Abstract:In this paper, we study the subspace clustering problem under complex noise. A wide class of reconstruction-based methods model the subspace clustering problem by combining a quadratic data-fidelity term and a regularization term. In a statistical framework, the data-fidelity term assumes to be contaminated by a unimodal Gaussian noise, which is a popular setting in most current subspace clustering models. However, the realistic noise is much more complex than our assumptions. Besides, the coarse representation of the data-fidelity term may depress the clustering accuracy, which is often used to evaluate the models. To address this issue, we propose the mixture of Gaussian regression (MoG Regression) for subspace clustering. The MoG Regression seeks a valid way to model the unknown noise distribution, which approaches the real one as far as possible, so that the desired affinity matrix is better at characterizing the structure of data in the real world, and furthermore, improving the performance. Theoretically, the proposed model enjoys the grouping effect, which encourages the coefficients of highly correlated points are nearly equal. Drawing upon the ideal of the minimum message length, a model selection strategy is proposed to estimate the numbers of the Gaussian components that shows a way how to seek the number of Gaussian components besides determining it by empirical value. In addition, the asymptotic property of our model is investigated. The proposed model is evaluated on the challenging datasets. The experimental results show that the proposed MoG Regression model significantly outperforms several state-of-the-art subspace clustering methods.

Noise-Aware Subspace Clustering

Subspace Clustering under Complex Noise

Subspace Clustering By Mixture Of Gaussian Regression

A Noise-Resistant Clustering Algorithm for Switching Regression Models

Model Inconsistent but Correlated Noise: Multi-view Subspace Learning with Regularized Mixture of Gaussians

Robust sparse regression by modeling noise as a mixture of gaussians

Correntropy Induced L2 Graph for Robust Subspace Clustering

Robust Subspace Clustering by Cauchy Loss Function

Robust Subspace Clustering Via Thresholding Ridge Regression

Unknown noise removal via sparse representation model

Learning Markov Random Walks for Robust Subspace Clustering and Estimation

Correntropy Metric-Based Robust Low-Rank Subspace Clustering for Motion Segmentation.

Robust deep multi-view subspace clustering networks with a correntropy-induced metric

Robust subspace clustering based on automatic weighted multiple kernel learning

Sparse Subspace Clustering Using Square-Root Penalty

Subspace Clustering by Capped Norm

Robust Least Squares Regression for Subspace Clustering: A Multi-View Clustering Perspective

Multiple Kernel Clustering with Dual Noise Minimization

Sparseness and Correntropy-Based Block Diagonal Representation for Robust Subspace Clustering

Matrix Factorization with Column L0-Norm Constraint for Robust Multi-subspace Analysis

Minimax Supervised Clustering in the Anisotropic Gaussian Mixture Model: A new take on Robust Interpolation