Benchmark RadarAE-UAV
AE-UAV is an air-to-air event-based UAV tracking benchmark with 178 flight sequences and continuous-time cubic B-spline annotations, supporting evaluation at arbitrary temporal resolutions. It includes multimodal auxiliary data and predefined train/validation/test splits.
- Released
- 2026-07-16
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
This benchmark addresses the lack of dedicated event-based datasets for air-to-air UAV tracking, enabling consistent evaluation and comparison of tracking methods under diverse motion and illumination conditions. It provides a public resource for developing and validating real-time tracking solutions on resource-constrained platforms.
Motivation
Air-to-air (A2A) unmanned aerial vehicle (UAV) tracking is fundamental to airborne remote sensing of low-altitude aerial targets.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.