Performance Analysis of a Foreground Segmentation Neural Network Model

TL;DR

This paper conducts an ablation study of FgSegNet_v2, proposing a variation that surpasses state-of-the-art performance in foreground segmentation across diverse datasets and conditions.

Contribution

It introduces a modified version of FgSegNet_v2 with improved accuracy, validated through extensive testing on multiple datasets.

Findings

Outperforms state-of-the-art in CDNet2014, especially in LowFrameRate subset

Achieves comparable results to state-of-the-art on SBI2015 and CityScapes datasets

Demonstrates robustness across different lighting and environmental conditions

Abstract

In recent years the interest in segmentation has been growing, being used in a wide range of applications such as fraud detection, anomaly detection in public health and intrusion detection. We present an ablation study of FgSegNet_v2, analysing its three stages: (i) Encoder, (ii) Feature Pooling Module and (iii) Decoder. The result of this study is a proposal of a variation of the aforementioned method that surpasses state of the art results. Three datasets are used for testing: CDNet2014, SBI2015 and CityScapes. In CDNet2014 we got an overall improvement compared to the state of the art, mainly in the LowFrameRate subset. The presented approach is promising as it produces comparable results with the state of the art (SBI2015 and Cityscapes datasets) in very different conditions, such as different lighting conditions.