HumiMimic: Mimetic Embodiment Transfer of Human-Humanoid Interaction from In-the
Overview
This research proposes HumiTransfer, a framework that converts in-the-wild human-human interaction videos into physically plausible human-humanoid interactions. The method reconstructs 3D human motion, transfers participants to target humanoid embodiments using interaction modeling, and produces a dataset of 500,000 frames for two humanoid robots. The approach preserves inter-body relationships and contact accuracy, with real-world deployments confirming the physical executability of transferred motions.
From arXiv Robotics — research abstracts
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · frames: 500000 other · Basis not reported
“e directions and anatomical offsets to prioritize interaction-critical inter-body relationships, while post-retargeting optimization enforces balance, contact, and collision avoidance. Subsequently, we apply HumiTransfer to 1,200 in-the-wild videos, producing approximately 500,000 frames of diverse interactions for the 26-DoF Unitree R1 Edu and the 23-DoF HighTorque Mini-hi humanoid robot, resulting in our HumiMimic”
Exact source · revision 2Source owner not reported
Reported quantity · videos: 1200 other · Basis not reported
“e directions and anatomical offsets to prioritize interaction-critical inter-body relationships, while post-retargeting optimization enforces balance, contact, and collision avoidance. Subsequently, we apply HumiTransfer to 1,200 in-the-wild videos, producing approximately 500,000 frames of diverse interactions for the 26-DoF Unitree R1 Edu and the 23-DoF HighTorque Mini-hi humanoid robot, resulting in our HumiMimic”
Exact source · revision 2Source owner not reported
Reported quantity · degrees of freedom: 26 other · Basis not reported · Differing source assertions
“e directions and anatomical offsets to prioritize interaction-critical inter-body relationships, while post-retargeting optimization enforces balance, contact, and collision avoidance. Subsequently, we apply HumiTransfer to 1,200 in-the-wild videos, producing approximately 500,000 frames of diverse interactions for the 26-DoF Unitree R1 Edu and the 23-DoF HighTorque Mini-hi humanoid robot, resulting in our HumiMimic”
Exact source · revision 2Source owner not reported
Reported quantity · degrees of freedom: 23 other · Basis not reported · Differing source assertions
“e directions and anatomical offsets to prioritize interaction-critical inter-body relationships, while post-retargeting optimization enforces balance, contact, and collision avoidance. Subsequently, we apply HumiTransfer to 1,200 in-the-wild videos, producing approximately 500,000 frames of diverse interactions for the 26-DoF Unitree R1 Edu and the 23-DoF HighTorque Mini-hi humanoid robot, resulting in our HumiMimic”
Exact source · revision 2Source owner not reported
Evidence dependency changes (1)
- 2026-10-31T05:04:01.523Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalHumiMimic: Mimetic Embodiment Transfer of Human-Humanoid Interaction from In-the-Wild Videos
This research proposes HumiTransfer, a framework that converts in-the-wild human-human interaction videos into physically plausible human-humanoid interactions. The method reconstructs 3D human motion, transfers participants to target humanoid embodiments using interaction modeling, and produces a dataset of 500,000 frames for two humanoid robots. The approach preserves inter-body relationships and contact accuracy, with real-world deployments confirming the physical executability of transferred motions.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).