1 paper
Jielin Wu, Shenzhe Yao, Guanqi He +6
Human hand-object demonstrations provide dense reference motions for training dexterous manipulation reinforcement learning (RL) policies through reference tracking. However, to us…