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

2025-06-04

AI Summary

Conventional RGB cameras suffer from motion blur, scale variation, and insufficient robustness when monitoring fast-moving, small-scale, and unpredictably maneuvering aerial objects (e.g., insects, birds, drones). To address these limitations, this work proposes a novel event-camera-based paradigm for detection and identification of flying objects. We introduce EV-Flying—the first large-scale outdoor event dataset featuring spatiotemporal bounding boxes and trajectory IDs. We design a lightweight, point-cloud-driven event representation that directly encodes asynchronous event streams as spatiotemporal point clouds, eliminating frame-rate constraints. A PointNet-inspired architecture enables end-to-end fine-grained classification and multi-object tracking. Experiments demonstrate substantial improvements in accuracy, robustness, and real-time performance for small and highly agile targets. Our approach establishes a scalable, event-driven framework for dynamic aerial biological monitoring.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Time-Series/Data Streams

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

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.

Problem

• Detect and recognize flying objects using event-based vision
• Address challenges like scale variations and motion blur
• Classify flying objects with point cloud event representations

Innovation

• Event-based vision for flying object detection
• Point-based approach with lightweight PointNet
• EV-Flying dataset with spatio-temporal annotations