Reproduce “Transformer IMU Calibrator: Dynamic On-body IMU Calibration for Inertial Motion Capture” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Chengxu Zuo, Jiawei Huang, Xiao Jiang, Yuan Yao, Xiangren Shi, Rui Cao, Xinyu Yi, Feng Xu and 2 more
Researchers from Xiamen University, Tsinghua University, Bournemouth University, and Cardiff University developed the Transformer IMU Calibrator (TIC), a method that dynamically calibrates Inertial Measurement Units during active motion. This system continuously estimates coordinate drift and sensor-to-body segment offset, resulting in significantly improved long-term accuracy and robustness for sparse inertial motion capture.
Key insights
- Accurate IMU calibration parameters can be dynamically estimated from active human motion by relaxing the global static assumption to a short-term static and diversity assumption.
- Leveraging a Transformer-based architecture allows for effective temporal modeling of IMU sequences to predict time-varying calibration parameters.
- Including acceleration data as an auxiliary input is crucial for distinguishing between global coordinate drift and sensor-to-body segment offset, enhancing parameter estimation accuracy.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.