Maritime Dataset for Distance Estimation
收藏资源简介:
该数据集由图宾根大学认知系统组创建,旨在解决无人水面车辆(USV)在海洋环境中的距离估计问题。数据集包含1000张图像,涵盖了开放水域、港口和沿海区域等多种海洋环境。每张图像都标注了船只、浮标等物体的边界框及其与USV的距离。数据集的创建过程包括使用USV的GPS位置和航向信息,结合NOAA的浮标数据,手动验证静态物体的距离。该数据集的应用领域主要集中于海洋辅助系统,旨在通过视觉线索进行距离估计,减少对昂贵传感器的依赖,提升USV的自主导航能力。
This dataset was developed by the Cognitive Systems Group at the University of Tübingen, with the core objective of addressing the distance estimation task for unmanned surface vehicles (USVs) in marine environments. The dataset consists of 1000 images covering diverse marine scenarios including open waters, ports, and coastal regions. Each image is annotated with bounding boxes for objects such as ships and buoys, as well as the distance between each annotated object and the USV. The dataset construction workflow leverages the GPS position and heading data of the USV, combined with buoy data from NOAA, to manually validate the distances of static objects. The primary application scope of this dataset lies in marine assistance systems, where it aims to enable distance estimation based on visual cues, reduce dependence on high-cost sensors, and enhance the autonomous navigation capability of USVs.

- 1Approximate Supervised Object Distance Estimation on Unmanned Surface Vehicles图宾根大学认知系统组 · 2025年



