MagShield: Towards Better Robustness in Sparse Inertial Motion Capture Under Magnetic Disturbances

Yunzhe Shao1 Xinyu Yi1 Lu Yin2 Shihui Guo2 Junhai Yong1 Feng Xu1

1 School of Software and BNRist, Tsinghua University

2 School of Informatics, Xiamen University

Accepted by ICCV 2025

We propose MagShield, an IMU orientation estimation method that specifically designed to address magnetic disturbances in sparse inertial MoCap systems. By integrating with our method, existing sparse inertial MoCap systems (left, PNP; right, DynaIP) can operate reliably even under severe magnetic condition.

Paper Supplementary Code Dataset Video Poster

Abstract

This paper proposes a novel method, named MagShield, designed to address the issue of magnetic disturbances in sparse inertial motion capture (MoCap) systems. Existing Inertial Measurement Units (IMUs) are prone to orienta tion estimation errors in magnetically disturbed environ ments, limiting the practical application of inertial Mocap systems in real-world scenarios. To address this problem, MagShield employs a “detect-then-correct” strategy, first detecting magnetic disturbances through multi-IMU joint analysis, and then correcting orientation errors using hu man motion priors. MagShield can be integrated with most existing sparse inertial MoCap systems, improving their performance in magnetically disturbed environments. Experimental results demonstrate that MagShield significantly enhances the accuracy of motion capture under magnetic interference and exhibits good compatibility across differ ent sparse inertial MoCap systems.

Live Demo

Comparison With ESKF

Comparison With Commercial Systems

In-The-Wild Performance

Potential Application

Acknowledgements

The authors would like to thank Siyuan Chang, Shao hua Pan, Wangyingqi Ding for their assistance with the live demos, and the volunteers who helped collect the MagIMU dataset. This work was supported by the National Key R&D Program of China (2023YFC3305600), the Zhejiang Provincial Natural Science Foundation (LDT23F02024F02), and the NSFC (No.61822111, 62021002). This work was also supported by THUIBCS, Tsinghua University, and BLBCI, Beijing Municipal Education Commission. Feng Xu is the corresponding author.