Dynamic Background Subtraction by Generative Neural Networks

TL;DR

This paper introduces DBSGen, a novel unsupervised generative neural network framework for dynamic background subtraction, effectively handling stochastic background movements in real-time computer vision applications.

Contribution

It proposes a unified, end-to-end trainable neural network model that improves dynamic background subtraction by separating motion removal and background generation tasks.

Findings

Outperforms most state-of-the-art methods on dynamic background sequences

Operates in an unsupervised and end-to-end manner

Effective in real-world applications with stochastic background movements

Abstract

Background subtraction is a significant task in computer vision and an essential step for many real world applications. One of the challenges for background subtraction methods is dynamic background, which constitute stochastic movements in some parts of the background. In this paper, we have proposed a new background subtraction method, called DBSGen, which uses two generative neural networks, one for dynamic motion removal and another for background generation. At the end, the foreground moving objects are obtained by a pixel-wise distance threshold based on a dynamic entropy map. The proposed method has a unified framework that can be optimized in an end-to-end and unsupervised fashion. The performance of the method is evaluated over dynamic background sequences and it outperforms most of state-of-the-art methods. Our code is publicly available at https://github.com/FatemeBahri/DBSGen.