Event-Aided Sharp Radiance Field Reconstruction for Fast-Flying Drones

TL;DR

This paper introduces a novel framework that combines asynchronous event streams and motion-blurred frames to enable high-fidelity radiance field reconstruction from fast-moving drones, overcoming challenges of motion blur and pose noise.

Contribution

It presents a unified event-image fusion approach integrated into NeRF optimization, jointly refining visual-inertial odometry without ground-truth supervision for agile drone imaging.

Findings

Reconstructs sharp radiance fields despite severe motion blur.

Achieves over 50% performance improvement on real-world drone data.

Successfully recovers accurate camera trajectories without ground-truth data.

Abstract

Fast-flying aerial robots promise rapid inspection under limited battery constraints, with direct applications in infrastructure inspection, terrain exploration, and search and rescue. However, high speeds lead to severe motion blur in images and induce significant drift and noise in pose estimates, making dense 3D reconstruction with Neural Radiance Fields (NeRFs) particularly challenging due to their high sensitivity to such degradations. In this work, we present a unified framework that leverages asynchronous event streams alongside motion-blurred frames to reconstruct high-fidelity radiance fields from agile drone flights. By embedding event-image fusion into NeRF optimization and jointly refining event-based visual-inertial odometry priors using both event and frame modalities, our method recovers sharp radiance fields and accurate camera trajectories without ground-truth supervision. We validate our approach on both synthetic data and real-world sequences captured by a fast-flying drone. Despite highly dynamic drone flights, where RGB frames are severely degraded by motion blur and pose priors become unreliable, our method reconstructs high-fidelity radiance fields and preserves fine scene details, delivering a performance gain of over 50% on real-world data compared to state-of-the-art methods.