Online and Batch Supervised Background Estimation via L1 Regression

TL;DR

This paper introduces a scalable L1 regression-based model for supervised video background estimation, demonstrating superior performance over existing methods in high-resolution scenarios through various efficient algorithms.

Contribution

The paper presents a simple, scalable L1 regression framework for background estimation, along with novel algorithms like reweighted least squares, homotopy, and stochastic gradient descent.

Findings

Outperforms state-of-the-art methods in accuracy

Effective for high-resolution videos

Multiple scalable algorithms demonstrated

Abstract

We propose a surprisingly simple model for supervised video background estimation. Our model is based on $\ell_1$ regression. As existing methods for $\ell_1$ regression do not scale to high-resolution videos, we propose several simple and scalable methods for solving the problem, including iteratively reweighted least squares, a homotopy method, and stochastic gradient descent. We show through extensive experiments that our model and methods match or outperform the state-of-the-art online and batch methods in virtually all quantitative and qualitative measures.