HORA

HORA: Hand–Object to Robot Action Dataset Dataset Summary HORA (Hand–Object to Robot Action) is a large-scale multimodal dataset that converts human hand–object interaction (HOI) demonstrations into robot-usable supervision for cross-embodiment learning. It combines HOI-style annotations (e.g., MANO hand parameters, object pose, contact) with embodied-robot learning signals (e.g., robot observations, end-effector trajectories) under a unified canonical action space. HORA… See the full description on the dataset page:

Source: Hugging Face Hub (HORA-DB/HORA). Metadata imported from the dataset’s Hub tags.