EV-Flying: an Event-based Dataset for In-The-Wild Recognition of Flying Objects

TL;DR

This paper introduces EV-Flying, a novel event-based dataset and method for recognizing flying objects like birds, insects, and drones, leveraging high-temporal-resolution event cameras and point-based neural architectures for improved aerial object detection.

Contribution

The paper presents the first event-based dataset for flying objects and demonstrates a point cloud-based approach for their recognition, addressing challenges of motion blur and high-speed movement.

Findings

Event cameras provide high temporal resolution for aerial detection.

PointNet-inspired architectures effectively classify flying objects.

EV-Flying dataset enables future research in real-world aerial recognition.

Abstract

Monitoring aerial objects is crucial for security, wildlife conservation, and environmental studies. Traditional RGB-based approaches struggle with challenges such as scale variations, motion blur, and high-speed object movements, especially for small flying entities like insects and drones. In this work, we explore the potential of event-based vision for detecting and recognizing flying objects, in particular animals that may not follow short and long-term predictable patters. Event cameras offer high temporal resolution, low latency, and robustness to motion blur, making them well-suited for this task. We introduce EV-Flying, an event-based dataset of flying objects, comprising manually annotated birds, insects and drones with spatio-temporal bounding boxes and track identities. To effectively process the asynchronous event streams, we employ a point-based approach leveraging lightweight architectures inspired by PointNet. Our study investigates the classification of flying objects using point cloud-based event representations. The proposed dataset and methodology pave the way for more efficient and reliable aerial object recognition in real-world scenarios.