This article introduces a large-scale dataset for 6-DoF localization of Unmanned Aerial Vehicles (UAVs) called UAVD4L. The dataset includes a 3D reference textured model which allows the generation of synthetic data such as rendered RGB and depth images. The authors also develop an offline-online pipeline for performing 6-DoF localization of UAVs in GPS-denied environments. The novel dataset covers approximately 2.5 million square meters, reconstructed from aerial oblique photography. The researchers believe this dataset and the two-stage localization pipeline will facilitate research in UAV localization in GPS-denied environments.
Publication date: 11 Jan 2024
Project Page: https://github.com/RingoWRW/UAVD4L
Paper: https://arxiv.org/pdf/2401.05971
