发表机构
OpenDriveLab at The University of Hong Kong; Alibaba Group; Shanghai Innovation Institute; Fudan University; Beihang University; Sichuan University; Archon Robotics（香港大学OpenDriveLab; 阿里巴巴集团; 上海创新研究院; 复旦大学; 北京航空航天大学; 四川大学; Archon机器人公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出首个以自我为中心的人到人形机器人全身协调移动操作技能迁移框架，通过粗到细动作对齐和视觉增强，实现零样本迁移，性能媲美遥操作且成本更低。
AI 中文摘要
人类演示能够捕捉多样的场景和丰富的全身技能，而无需机器人遥操作。先前的以自我为中心（egocentric）的迁移工作强调在解耦控制下移动操作中的场景泛化，而对协调全身技能的直接迁移探索较少。我们提出EgoHumanoid-V2，这是首个用于协调全身移动操作的以自我为中心的人到人形机器人技能迁移框架。其核心是粗到细的动作对齐，结合了运动学参考校正与动力学感知细化。该方法提高了末端执行器姿态精度，同时保持了全身协调性。我们还使用机器人手臂渲染和训练时图像增强来缩小视觉具身差距并提高视角鲁棒性。在四个真实世界任务中，基于对齐人类数据训练的语言-视觉-动作（VLA）策略展示了零样本技能迁移，无需目标任务的机器人演示。任务得分与基于遥操作数据训练的策略相当，且采集成本更低。这些结果支持人类数据可直接作为技能监督。
英文摘要
Human demonstrations capture diverse scenes and rich whole-body skills without requiring robot teleoperation. Prior work on egocentric transfer has emphasized scene generalization in loco-manipulation under decoupled control, leaving direct transfer of coordinated whole-body skills less explored. We present EgoHumanoid-V2, the first egocentric human-to-humanoid skill transfer framework for coordinated whole-body loco-manipulation. At its core, coarse-to-fine action alignment combines kinematic reference correction with dynamics-aware refinement. It improves end-effector pose accuracy while preserving whole-body coordination. We also use robot-arm rendering and training-time image augmentation to reduce the visual embodiment gap and improve viewpoint robustness. On four real-world tasks, vision-language-action (VLA) policies trained on aligned human data show zero-shot skill transfer without target-task robot demonstrations. Task scores are comparable to those of policies trained on teleoperation data at a lower collection cost. These results support human data as direct skill supervision.
