C2FDrone: Coarse-to-Fine Drone-to-Drone Detection using Vision Transformer Networks

TL;DR

This paper introduces C2FDrone, a vision transformer-based detection system that improves drone-to-drone detection accuracy and real-time performance, addressing challenges like small object size and occlusion.

Contribution

The paper presents a novel coarse-to-fine detection approach using vision transformers, enhancing detection accuracy on challenging datasets and enabling real-time deployment on edge devices.

Findings

F1 score improved by up to 7% on FL-Drones dataset

Achieved real-time processing on edge hardware

Demonstrated robustness against blur and small objects

Abstract

A vision-based drone-to-drone detection system is crucial for various applications like collision avoidance, countering hostile drones, and search-and-rescue operations. However, detecting drones presents unique challenges, including small object sizes, distortion, occlusion, and real-time processing requirements. Current methods integrating multi-scale feature fusion and temporal information have limitations in handling extreme blur and minuscule objects. To address this, we propose a novel coarse-to-fine detection strategy based on vision transformers. We evaluate our approach on three challenging drone-to-drone detection datasets, achieving F1 score enhancements of 7%, 3%, and 1% on the FL-Drones, AOT, and NPS-Drones datasets, respectively. Additionally, we demonstrate real-time processing capabilities by deploying our model on an edge-computing device. Our code will be made publicly available.