ReContact: Reconstruct and Retarget Human Contacts from Video for Contact-Rich Humanoid Manipulation
ReContact: Reconstruct and Retarget Human Contacts from Video for Contact-Rich Humanoid Manipulation
This paper introduces ReContact, a framework that reconstructs and retargets human-object interactions from video to enable contact-rich manipulation for humanoid robots. By defining an object-centric contact plan that specifies phases, pairs, anchors, directions, and modes, the method stabilizes pose reconstruction under occlusion and ensures forces align with planned contacts. In simulation, ReContact achieves 83.5% success in thigh-supported regrasp tasks compared to 0.16% for baselines, and improves tube-transfer success from 20.4% to 97.8% using a single policy trained on 247 references. On a real Unitree G1, the system executes seven contact-rich manipulation and two loco-manipulation tasks.
The study presents ReContact, a method that reconstructs and retargets human-object interactions from video to enable contact-rich manipulation on humanoid robots, achieving high success rates in simulated regrasp and tube-transfer tasks while executing seven manipulation and two loco-manipulation tasks on a real Unitree G1.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S4
Reported numbers
number of tasks
7
View original evidence
ed for new placements raises in-workspace tube-transfer success from 20.4% (source reference only) to 97.8%. On a real Unitree G1, policies execute seven contact-rich manipulation and two loco-manipulation tasks.
Open source S5number of tasks
2
View original evidence
ed for new placements raises in-workspace tube-transfer success from 20.4% (source reference only) to 97.8%. On a real Unitree G1, policies execute seven contact-rich manipulation and two loco-manipulation tasks.
Open source S5number of reference videos
247
View original evidence
econstruction through occlusion. A residual policy over a frozen whole-body tracker is rewarded for making contact at planned anchors with forces aligned to the specified directions. In simulation, on a thigh-supported regrasp task, ReContact achieves the planned two-palm bottom hold in 83.5 of episodes over three training seeds, versus 0.16% for the best adapted baseline. One policy trained on 247 references re-solv
Open source S4number of training seeds
>3
View original evidence
econstruction through occlusion. A residual policy over a frozen whole-body tracker is rewarded for making contact at planned anchors with forces aligned to the specified directions. In simulation, on a thigh-supported regrasp task, ReContact achieves the planned two-palm bottom hold in 83.5 of episodes over three training seeds, versus 0.16% for the best adapted baseline. One policy trained on 247 references re-solv
Open source S4
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
econstruction through occlusion. A residual policy over a frozen whole-body tracker is rewarded for making contact at planned anchors with forces aligned to the specified directions. In simulation, on a thigh-supported regrasp task, ReContact achieves the planned two-palm bottom hold in 83.5 of episodes over three training seeds, versus 0.16% for the best adapted baseline. One policy trained on 247 references re-solv
Open source S4
ed for new placements raises in-workspace tube-transfer success from 20.4% (source reference only) to 97.8%. On a real Unitree G1, policies execute seven contact-rich manipulation and two loco-manipulation tasks.
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org