Robust Matrix Factorization with Unknown Noise Deyu

Abstract

Many problems in computer vision can be posed as recovering a low-dimensional subspace from highdimensional visual data. Factorization approaches to lowrank subspace estimation minimize a loss function between an observed measurement matrix and a bilinear factorization. Most popular loss functions include the L2 and L1 losses. L2 is optimal for Gaussian… (More)

6 Figures and Tables

Topics

  • Presentations referencing similar topics