LAFAN-1

'LAFAN-1' is a dataset used to evaluate the performance of humanoid robots in retargeting human motions for loco-manipulation tasks, focusing on preserving interactions with objects and the environment.

Papers using LAFAN-1 (2)

RobotDancing: Residual-Action Reinforcement Learning Enables Robust Long-Horizon Humanoid Motion Tracking2025OmniRetarget: Interaction-Preserving Data Generation for Humanoid Whole-Body Loco-Manipulation and Scene Interaction2025