| title | HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone |
|---|---|
| authors | Simple AI, Yuteng Wei, Jinming Ma, Jiawei Wang, Weitao Zhou, Yushen Zuo, Ke Rui, Minglei Li, Jinhao Zhang, Zhikang Pan, Xiang Wang, Haoran Jia, Huan Du, Zicheng Zeng, Jun Ma, Guiyu Qin, Di Zhang, Xiaofei Li |
| date | 2026-07-28 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2607.25895 |
| affiliation | Simple World Lab |
| upvotes | 61 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf |
| abstract | Learning deployable manipulation policies is bottlenecked by the scarcity of data that is both high-fidelity and scalable. Real-robot teleoperation is accurate but costly to scale; robot-free UMI capture scales readily, and current practice uses the resulting data mainly for pre-training, adding a small real-robot "anchor" at post-training. We ask whether raising the fidelity of robot-free UMI data, rather than shrinking the real-robot fraction, can remove that anchor. We present HiFi-UMI, a portable UMI data-production system co-designed for trajectory accuracy, inter-gripper relative pose, synchronization, and field of view: head-mounted offline stereo-inertial SLAM, native rather than reconstructed relative pose, a shared microsecond GPIO trigger, and two wide-angle cameras per hand covering ~200 degrees. It reaches 3 mm workspace-local end-effector accuracy without external tracking infrastructure. Using this corpus, we demonstrate zero-robot post-training: a policy post-trained solely on HiFi-UMI demonstrations deploys directly on a real robot and matches in-domain teleoperation across three backbones spanning the vision-language-action and world-action-model families, with success-rate differences of -2.5, +3.1, and -0.6 percentage points on StarVLA-QwenPI, OpenPI-pi_0.5, and LingBot-VA; the strongest policy reaches 85% on a precision insertion task, even though the teleoperation baseline is collected in the evaluation scene and no HiFi-UMI trajectory is. Pre-training on 4,000 hours from the same corpus lowers action error on ten unseen tasks by 41% and, on StarVLA-QwenPI, raises real-robot success by a further 18.1 percentage points. We open-source HiFi-UMI-2K, 2,000 hours of microsecond-synchronized, ultra-wide-FoV demonstrations, each automatically reconstructed and validated through simulation replay, as a large-scale, high-fidelity resource for the robot-learning community. |
cite
@misc{ai2026hifi,
title={HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone},
author={Simple AI and Yuteng Wei and Jinming Ma and Jiawei Wang and Weitao Zhou and Yushen Zuo and Ke Rui and Minglei Li and Jinhao Zhang and Zhikang Pan and Xiang Wang and Haoran Jia and Huan Du and Zicheng Zeng and Jun Ma and Guiyu Qin and Di Zhang and Xiaofei Li},
year={2026},
eprint={2607.25895},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2607.25895}
}