arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.05971cs.CV
收藏

UAVD4L:用于无人机6自由度定位的大规模数据集

UAVD4L: A Large-Scale Dataset for UAV 6-DoF Localization

National University of Defense Technology(国防科技大学)

Rouwan Wu, Xiaoya Cheng, Juelin Zhu, Xuxiang Liu, Maojun Zhang, Shen Yan

AI总结:

针对GPS拒止环境下无人机定位数据集规模小、视角单一、缺乏真值和内置传感器数据的问题,本文提出大规模6自由度定位数据集UAVD4L及两阶段定位流程UAVLoc,并基于此构建三维地面目标跟踪系统,实验验证了其有效性。

摘要:

尽管在GPS拒止环境中无人机(UAV)的全局定位取得了显著进展,现有方法仍受限于数据集的可用性。当前数据集通常聚焦于小规模场景,缺乏视角多样性、精确的真值(GT)位姿以及无人机内置传感器数据。为解决这些局限性,我们引入了一个用于定位的大规模6自由度无人机数据集(UAVD4L),并开发了一个两阶段6自由度定位流程(UAVLoc),该流程由离线合成数据生成和在线视觉定位组成。此外,基于6自由度估计器,我们设计了一个用于在三维空间中跟踪地面目标的分层系统。在新数据集上的实验结果表明了所提方法的有效性。代码和数据集可在 https://github.com/RingoWRW/UAVD4L 获取。

英文摘要:

Despite significant progress in global localization of Unmanned Aerial Vehicles (UAVs) in GPS-denied environments, existing methods remain constrained by the availability of datasets. Current datasets often focus on small-scale scenes and lack viewpoint variability, accurate ground truth (GT) pose, and UAV build-in sensor data. To address these limitations, we introduce a large-scale 6-DoF UAV dataset for localization (UAVD4L) and develop a two-stage 6-DoF localization pipeline (UAVLoc), which consists of offline synthetic data generation and online visual localization. Additionally, based on the 6-DoF estimator, we design a hierarchical system for tracking ground target in 3D space. Experimental results on the new dataset demonstrate the effectiveness of the proposed approach. Code and dataset are available at https://github.com/RingoWRW/UAVD4L

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑