Artriy/FRED

FRED（佛罗伦萨RGB-事件无人机数据集）是一个专门设计用于无人机检测、跟踪和轨迹预测的大规模多模态数据集，包含时空同步的RGB和事件数据。数据集提供训练和测试分割，每个序列以压缩文件形式存储，包含RGB帧、事件数据（包括原始事件流和提取的事件帧）以及标注信息。标注文件采用扩展边界框格式，包含时间戳、边界框坐标、无人机唯一标识和类别标签。数据集涵盖多种场景，如夜间、雨天和室内环境，适用于标准检测和跟踪流程。FRED (Florence RGB-Event Drone dataset) is a large-scale multimodal dataset specifically designed for drone detection, tracking, and trajectory forecasting, with spatiotemporally synchronized RGB and event data. It includes train and test splits with zipped subfolders for each sequence, containing rgb frames, event data (both extracted event frames and raw event stream in .hdf5 files), and annotations. The annotation format uses extended bounding boxes with timestamps, coordinates, unique drone identifiers, and class labels. The dataset features diverse scenarios such as night, raining, and indoor environments, and is compatible with standard detection and tracking pipelines.