UAVDT 无人机目标检测追踪视频数据集
收藏资源简介:
UAVDT 全称 Unmanned Aerial Vehicle Benchmark Object Detection and Tracking,是关于无人机目标检测追踪的大规模视频数据集,包含 10 小时的原始视频,约 8,000 具有代表性的视频帧,这些视频帧带有手动标注边界框和一些有用的标签,例如车辆类别和遮挡。该数据集是由无人机在各种复杂场景下捕获的,主要用于执行三个基本任务:目标检测(DET)、单目标跟踪(SOT)和多目标跟踪(MOT)。
UAVDT, short for Unmanned Aerial Vehicle Benchmark Object Detection and Tracking, is a large-scale video dataset dedicated to unmanned aerial vehicle (UAV)-based object detection and tracking. It contains 10 hours of raw video footage and approximately 8,000 representative video frames, all manually annotated with bounding boxes and informative labels such as vehicle categories and occlusion status. Captured by drones in various complex scenarios, this dataset is primarily designed for three fundamental tasks: object detection (DET), single-object tracking (SOT) and multi-object tracking (MOT).




