An Online Algorithm for Separating Sparse and Low-dimensional Signal Sequences from their Sum

Abstract

This paper designs and evaluates a practical algorithm, called practical recursive projected compressive sensing (Prac-ReProCS), for recovering a time sequence of sparse vectors $S_t$ and a time sequence of dense vectors $L_t$ from their sum, $M_t:=S_t+L_t$, when any subsequence of the $L_t$'s lies in a slowly changing low-dimensional subspace. A key application where this problem occurs is in video layering where the goal is to separate a video sequence into a slowly changing background sequence and a sparse foreground sequence that consists of one or more moving regions/objects. Prac-ReProCS is a practical modification of its theoretical counterpart which was analyzed in our recent work. Experimental comparisons demonstrating the advantage of the approach for both simulated and real videos are shown. Extension to the undersampled case is also developed.

Accepted by IEEE Transactions on Signal Processing. The title has been changed from "Practical ReProCS for Separating Sparse and Low-dimensional Signal Sequences from their Sum" to "An Online Algorithm for Separating Sparse and Low-dimensional Signal Sequences from their Sum"

References in corpus (8)

• Dynamic Anomalography: Tracking Network Anomalies via Sparsity and Low Rank
• Estimation of Simultaneously Sparse and Low Rank Matrices
• Recovery of Low-Rank Plus Compressed Sparse Matrices with Application to Unveiling Traffic Anomalies
• ReProCS: A Missing Link between Recursive Robust PCA and Recursive Sparse Recovery in Large but Correlated Noise
• Weighted ${\ell }_1$ Minimization for Sparse Recovery with Prior Information
• Sharp recovery bounds for convex demixing, with applications
• Robust PCA and subspace tracking from incomplete observations using L0-surrogates
• Recursive Sparse Recovery in Large but Structured Noise - Part 2

Cited by in corpus (24)

• Robust Subspace Learning: Robust PCA, Robust Subspace Tracking, and Robust Subspace Recovery
• Image Reconstruction: From Sparsity to Data-adaptive Methods and Machine Learning
• Decomposition into Low-rank plus Additive Matrices for Background/Foreground Separation: A Review for a Comparative Evaluation with a Large-Scale Dataset
• Total Variation Regularized Tensor RPCA for Background Subtraction from Compressive Measurements
• Recursive Recovery of Sparse Signal Sequences from Compressive Measurements: A Review
• High Dimensional Low Rank plus Sparse Matrix Decomposition
• Low-rank and Adaptive Sparse Signal (LASSI) Models for Highly Accelerated Dynamic Imaging
• Static and Dynamic Robust PCA and Matrix Completion: A Review
• Robust PCA with Partial Subspace Knowledge
• Provable Dynamic Robust PCA or Robust Subspace Tracking
• On the Adversarial Robustness of Subspace Learning
• Video Denoising and Enhancement via Dynamic Video Layering
• Fast Robust Subspace Tracking via PCA in Sparse Data-Dependent Noise
• Deep Learning for Biomedical Image Reconstruction: A Survey
• Weighted Low-Rank Approximation of Matrices and Background Modeling
• Incorporating Prior Information in Compressive Online Robust Principal Component Analysis
• Online (and Offline) Robust PCA: Novel Algorithms and Performance Guarantees
• Bridging Convex and Nonconvex Optimization in Robust PCA: Noise, Outliers, and Missing Data
• A Batch-Incremental Video Background Estimation Model using Weighted Low-Rank Approximation of Matrices
• Online Decomposition of Compressive Streaming Data Using $n$-${\ell }_1$ Cluster-Weighted Minimization
• Low-Rank Methods in Event Detection and Subsampled Point-to-Subspace Proximity Tests
• A Correctness Result for Online Robust PCA
• Ocean Reverberation Suppression via Matrix Completion with Sensor Failure
• Robust Online Matrix Factorization for Dynamic Background Subtraction