RoSHI: A Versatile Robot-oriented Suit for Human Data In-the-Wild

TL;DR

RoSHI is a hybrid wearable system combining IMUs and egocentric glasses to accurately estimate full 3D human pose in real-world settings, facilitating robot learning and human data collection.

Contribution

The paper introduces RoSHI, a novel hybrid wearable system that fuses IMUs with egocentric perception to improve 3D human pose estimation in natural environments.

Findings

RoSHI outperforms other egocentric baselines in pose estimation.

RoSHI performs comparably to state-of-the-art exocentric methods.

Recorded motion data is effective for humanoid policy learning.

Abstract

Scaling up robot learning will likely require human data containing rich and long-horizon interactions in the wild. Existing approaches for collecting such data trade off portability, robustness to occlusion, and global consistency. We introduce RoSHI, a hybrid wearable that fuses low-cost sparse IMUs with the Project Aria glasses to estimate the full 3D pose and body shape of the wearer in a metric global coordinate frame from egocentric perception. This system is motivated by the complementarity of the two sensors: IMUs provide robustness to occlusions and high-speed motions, while egocentric SLAM anchors long-horizon motion and stabilizes upper body pose. We collect a dataset of agile activities to evaluate RoSHI. On this dataset, we generally outperform other egocentric baselines and perform comparably to a state-of-the-art exocentric baseline (SAM3D). Finally, we demonstrate that the motion data recorded from our system are suitable for real-world humanoid policy learning. For videos, data and more, visit the project webpage: https://roshi-mocap.github.io/