Deep Neural Network Concepts for Background Subtraction: A Systematic Review and Comparative Evaluation

Abstract

Conventional neural networks show a powerful framework for background subtraction in video acquired by static cameras. Indeed, the well-known SOBS method and its variants based on neural networks were the leader methods on the largescale CDnet 2012 dataset during a long time. Recently, convolutional neural networks which belong to deep learning methods were employed with success for background initialization, foreground detection and deep learned features. Currently, the top current background subtraction methods in CDnet 2014 are based on deep neural networks with a large gap of performance in comparison on the conventional unsupervised approaches based on multi-features or multi-cues strategies. Furthermore, a huge amount of papers was published since 2016 when Braham and Van Droogenbroeck published their first work on CNN applied to background subtraction providing a regular gain of performance. In this context, we provide the first review of deep neural network concepts in background subtraction for novices and experts in order to analyze this success and to provide further directions. For this, we first surveyed the methods used background initialization, background subtraction and deep learned features. Then, we discuss the adequacy of deep neural networks for background subtraction. Finally, experimental results are presented on the CDnet 2014 dataset.

arXiv comment46 pages, 4 figures, submitted to neural networks

References in corpus (13)

• Deep Learning in Neural Networks: An Overview
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Caffe: Convolutional Architecture for Fast Feature Embedding
• FlowNet: Learning Optical Flow with Convolutional Networks
• Rethinking the Inception Architecture for Computer Vision
• Foreground Segmentation Using a Triplet Convolutional Neural Network for Multiscale Feature Encoding
• Learning in Memristive Neural Network Architectures using Analog Backpropagation Circuits
• Fast Feature Fool: A data independent approach to universal adversarial perturbations
• Online Robust Subspace Tracking from Partial Information
• Mathematics of Deep Learning
• Combining Background Subtraction Algorithms with Convolutional Neural Network
• Joint Background Reconstruction and Foreground Segmentation via A Two-stage Convolutional Neural Network

Cited by in corpus (3)

• Graph CNN for Moving Object Detection in Complex Environments from Unseen Videos
• Background Subtraction with Real-time Semantic Segmentation
• Illumination Invariant Foreground Object Segmentation using ForeGANs