RL-Based Retargeting Method For Transferring Human Motion To Robots
Disney Research presented ReActor.
With a new video, Disney Research introduced ReActor, a research project designed to retarget human motion to robots with different sizes and body structures, including humanoids and quadrupeds. The method is said to enable the transfer of highly dynamic, contact-rich movements while minimizing common issues like foot sliding, hovering, and self-penetration.
The ReActor framework operates on two levels: it simultaneously adapts reference motions to match a robot's body structure while training a tracking policy through reinforcement learning. The method only needs a few basic matching points between the human and robot body parts, and it automatically finds the best settings.
ReActor is part of the company's ongoing work in robotics, an area it has been exploring for years. The abstract of the full research paper states:
"Retargeting human kinematic reference motion onto a robot's morphology remains a formidable challenge. Existing methods often produce physical inconsistencies, such as foot sliding, self-collisions, or dynamically infeasible motions, which hinder downstream imitation learning.
We propose a bilevel optimization framework that jointly adapts reference motions to a robot's morphology while training a tracking policy using reinforcement learning. To make the optimization tractable, we derive an approximate gradient for the upper-level loss. Our framework requires only a sparse set of semantic rigid-body correspondences and eliminates the need for manual tuning by identifying optimal values for a parameterization expressive enough to preserve characteristic motion across different embodiments.
Moreover, by integrating retargeting directly with physics simulation, we produce physically plausible motions that facilitate robust imitation learning. We validate our method in simulation and on hardware, demonstrating challenging motions for morphologies that differ significantly from a human, including retargeting onto a quadruped."
ReActor: Reinforcement Learning for Physics-Aware Motion Retargeting was developed by David Müller, Agon Serifi, Sammy Christen, Ruben Grandia, Espen Knoop, and Moritz Bächer. It may be presented at SIGGRAPH 2026, but no further details, including potential future applications, have been officially shared.
Don't forget to subscribe to our Newsletter, join our 80 Level Talent platform, and follow us on Twitter, LinkedIn, Telegram, and Instagram, where we share breakdowns, the latest news, awesome artworks, and more.
Are you a fan of what we do here at 80 Level? Then make sure to set us as a Preferred Source on Google to see more of our content in your feed.
