aiqbench analysis
RoboTryOn advances robot learning by treating human-to-robot task translation as a virtual try-on problem, effectively swapping human actors for robots while preserving motion dynamics. This approach leverages synthetic data derived from egocentric videos to train policies that achieve significantly higher success rates on novel manipulation tasks without requiring real-world demonstrations for every new skill. By reducing data collection costs and bridging the gap between high-degree-of-freedom robot embodiments and human demonstration sources, this method offers a scalable pathway for deploying foundation models in complex robotic environments.
RoboTryOn: Synthesizing Dexterous Robot Demonstrations from Human Videos via Virtual Try-On
arXiv:2611.05163v1 Announce Type: cross Abstract: Robot foundation models have shown remarkable performance in robot control, yet adapting them to a new task still relies heavily on teleoperated demonstrations. Egocentric human videos offer a scalable substitute, but overlaying rendered robots on human videos or retargeting human poses into robot actions leaves significant visual or action gaps for high-DoF robot embodiments. To address this limitation, we propose RoboTryOn, which treats human-to-robot translation as virtual try-on: try-on keeps the person and swaps the garment, RoboTryOn keeps the demonstrated motion and swaps the human for the robot. Specifically, RoboTryOn converts a human video into a video of the target robot specified by a reference image, while preserving the original task, objects, and background. To maintain visual consistency, we translate selected keyframes autoregressively and use them to guide video synthesis with motion conditioned on the human hand skeleton. Additionally, we use an inverse dynamics model (IDM) that incorporates depth features to infer action labels from synthesized videos for policy learning. Our experiments across multiple humanoid embodiments show that policies trained with our synthetic data improve the average success rate on novel tasks, including challenging bimanual manipulation, from 2% to 60%, without any real robot demonstration of those tasks. Moreover, training with our synthetic data matches the performance of policies trained solely on teleoperated demonstrations at approximately a third of the collection cost, demonstrating the cost efficiency of our approach.
Abstract:Robot foundation models have shown remarkable performance in robot control, yet adapting them to a new task still relies heavily on teleoperated demonstrations. Egocentric human videos offer a scalable substitute, but overlaying rendered robots on human videos or retargeting human poses into robot actions leaves significant visual or action gaps for high-DoF robot embodiments. To address this limitation, we propose RoboTryOn, which treats human-to-robot translation as virtual try-on: try-on keeps the person and swaps the garment, RoboTryOn keeps the demonstrated motion and swaps the human for the robot. Specifically, RoboTryOn converts a human video into a video of the target robot specified by a reference image, while preserving the original task, objects, and background. To maintain visual consistency, we translate selected keyframes autoregressively and use them to guide video synthesis with motion conditioned on the human hand skeleton. Additionally, we use an inverse dynamics model (IDM) that incorporates depth features to infer action labels from synthesized videos for policy learning. Our experiments across multiple humanoid embodiments show that policies trained with our synthetic data improve the average success rate on novel tasks, including challenging bimanual manipulation, from 2% to 60%, without any real robot demonstration of those tasks. Moreover, training with our synthetic data matches the performance of policies trained solely on teleoperated demonstrations at approximately a third of the collection cost, demonstrating the cost efficiency of our approach.
29 pages, 15 figures, 13 tables. Project page: this https URL
Robotics (cs.RO); Artificial Intelligence (cs.AI)
arXiv:2611.05163 [cs.RO]
(or arXiv:2611.05163v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2611.05163 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 336 words
Related coverage
Demonstration Synthesis from a Single Scan via Gaussian Splatting for Visuomotor Policy Learning
arXiv · Sep 21, 2026
V2-STRep: VLM-Grounded Structured Task Representations for Reusable Robot Skills Acquired from Generated Videos
arXiv · Sep 18, 2026
SMART: Zero-Shot Sim-to-Real Articulated Object Manipulation via Large-Scale Synthetic Pretraining
arXiv · Oct 7, 2026
DynamicManip: Enabling Dynamic Manipulation from a Single Static Demonstration
arXiv · Aug 4, 2026