Robust Humanoid Motion Retargeting

Summary

This automated pipeline retargets diverse motion sources, including noisy pose estimates, into natural and physically feasible movements for many humanoid robots. It unifies different skeletons through a canonical rig, corrects infeasible poses and foot contact, and then refines trajectories within each robot’s physical limits. Tests on 12 simulated and three real robots produced expressive upper-body motion with consistent foot contact. (Jeong et al., 2025)

Contribution

I coauthored this work and contributed to the implementation required to execute the motion-retargeting code on physical robot platforms.

Resources

Project Page · IEEE Xplore · PDF · GitHub

References

2025

IROS

Robust and Expressive Humanoid Motion Retargeting via Optimization-Based Rig Unification

Taemoon Jeong, Taehyun Byun, Jihoon Kim, Keunjun Choi, Jaesung Oh, Sungpyo Lee, Omar Darwish, Joohyung Kim, and Sungjoon Choi

In 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2025

DOI Bib HTML PDF Code Website

@inproceedings{jeong2025robust,
  title = {Robust and Expressive Humanoid Motion Retargeting via Optimization-Based Rig Unification},
  author = {Jeong, Taemoon and Byun, Taehyun and Kim, Jihoon and Choi, Keunjun and Oh, Jaesung and Lee, Sungpyo and Darwish, Omar and Kim, Joohyung and Choi, Sungjoon},
  booktitle = {2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages = {21619--21626},
  year = {2025},
  doi = {10.1109/IROS60139.2025.11246607},
}