Firearm Detection via Convolutional Neural Networks: Comparing a Semantic Segmentation Model Against End-to-End Solutions

TL;DR

This paper compares a semantic segmentation neural network model to an end-to-end deep learning model for firearm detection in videos, highlighting trade-offs in accuracy, flexibility, and data efficiency.

Contribution

It introduces a comparison framework between semantic segmentation and end-to-end models for weapon detection, emphasizing the advantages of segmentation in low-data scenarios.

Findings

Semantic segmentation offers greater flexibility and resilience with limited data.

End-to-end models achieve higher accuracy but require more data and tuning.

Segmentation models are more adaptable to different configurations.

Abstract

Threat detection of weapons and aggressive behavior from live video can be used for rapid detection and prevention of potentially deadly incidents such as terrorism, general criminal offences, or even domestic violence. One way for achieving this is through the use of artificial intelligence and, in particular, machine learning for image analysis. In this paper we conduct a comparison between a traditional monolithic end-to-end deep learning model and a previously proposed model based on an ensemble of simpler neural networks detecting fire-weapons via semantic segmentation. We evaluated both models from different points of view, including accuracy, computational and data complexity, flexibility and reliability. Our results show that a semantic segmentation model provides considerable amount of flexibility and resilience in the low data environment compared to classical deep model models, although its configuration and tuning presents a challenge in achieving the same levels of accuracy as an end-to-end model.