A Deep-Unfolded Reference-Based RPCA Network For Video Foreground-Background Separation

TL;DR

This paper introduces a deep-unfolded neural network for video foreground-background separation that models temporal correlations, leading to faster convergence and higher accuracy than existing methods.

Contribution

It presents a novel deep-unfolding approach for RPCA that incorporates temporal correlation modeling and adaptive proximal operators, improving performance in video separation tasks.

Findings

Outperforms state-of-the-art RPCA networks on moving MNIST dataset.

Achieves faster convergence and higher accuracy.

Effectively models temporal correlations in video frames.

Abstract

Deep unfolded neural networks are designed by unrolling the iterations of optimization algorithms. They can be shown to achieve faster convergence and higher accuracy than their optimization counterparts. This paper proposes a new deep-unfolding-based network design for the problem of Robust Principal Component Analysis (RPCA) with application to video foreground-background separation. Unlike existing designs, our approach focuses on modeling the temporal correlation between the sparse representations of consecutive video frames. To this end, we perform the unfolding of an iterative algorithm for solving reweighted $\ell_1$-$\ell_1$ minimization; this unfolding leads to a different proximal operator (a.k.a. different activation function) adaptively learned per neuron. Experimentation using the moving MNIST dataset shows that the proposed network outperforms a recently proposed state-of-the-art RPCA network in the task of video foreground-background separation.