The DARRL dataset: Demonstrations for Action Recognition and Robot Learning

The DARRL dataset (Demonstrations for Action Recognition and Robot Learning) is a collection of 760 RGB-D videos of humans performing various manipulation tasks. It is provided with object and action annotations (in the COCO format) for 30 of those videos; segmentation masks are also provided. Additionnal gaze data collected from outside observers is shared for research on human attention and observation strategy analysis. It can also be used as a basis for learning from demonstrations for a robotic arm, for instance. This work is supported by Région Pays de la Loire.DARRL数据集（Demonstrations for Action Recognition and Robot Learning）是一个收录760段人类执行各类操作任务的RGB-D视频的数据集。该数据集为其中30段视频提供了COCO格式的目标与动作标注，同时附带分割掩码。此外，数据集还共享了由外部观察者采集的注视数据，可用于人类注意力与观察策略分析相关研究。 例如，该数据集也可作为机械臂从演示中学习的基础。 本研究得到了卢瓦尔河地区大区（Région Pays de la Loire）的资助。