Research map / 3D vision / Human pose estimation
MagShield: Towards Better Robustness in Sparse Inertial Motion Capture Under Magnetic Disturbances
Yunzhe Shao, Xinyu Yi, Lu Yin, Shihui Guo, Junhai Yong, Feng Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add multi-IMU pose-aware magnetic disturbance detector to baseline (ablation) | PNP · MagIMU (local pose) | angular error (deg): 24.65 -> 23.20 | Improved |
| add YawCorrector net on top of baseline+detector (ablation) | PNP · MagIMU (local pose) | angular error (deg): 23.20 -> 20.23 | Improved |
| add multi-IMU pose-aware magnetic disturbance detector to baseline (ablation) | DynaIP · MagIMU (local pose) | angular error (deg): 28.79 -> 27.26 | Improved |
| add YawCorrector net on top of baseline+detector (ablation) | DynaIP · MagIMU (local pose) | angular error (deg): 27.26 -> 22.12 | Improved |
| add detector and net (full MagShield) to baseline | PNP · MagIMU (local pose) | SIP error (deg): 26.63 -> 24.19 | Improved |
| add detector and net (full MagShield) to baseline | DynaIP · MagIMU (local pose) | SIP error (deg): 31.55 -> 28.68 | Improved |
| add detector and net (full MagShield) to baseline | PNP · MagIMU (local pose) | positional error (cm): 8.99 -> 8.17 | Improved |
| add detector and net (full MagShield) to baseline | DynaIP · MagIMU (local pose) | mesh error (cm): 11.17 -> 9.84 | Improved |
| synthesize training IMU data by simulating magnetic field noise instead of adding yaw noise to clean readings (ablation) | PNP · MagIMU (local pose) | angular error (deg): 20.53 -> 20.23 | Improved |
| synthesize training IMU data by simulating magnetic field noise instead of adding yaw noise to clean readings (ablation) | DynaIP · MagIMU (local pose) | angular error (deg): 22.51 -> 22.12 | Improved |
| integrate MagShield on clean-field TotalCapture | PNP · TotalCapture (clean magnetic field) | pose error | No clear effect |
Closest papers by meaning
- Dynamic Inertial Poser (DynaIP): Part-Based Motion Dynamics Learning for Enhanced Human Pose Estimation with Sparse Inertial Sensors (CVPR 2024 · similarity 0.80)
- iMoT: Inertial Motion Transformer for Inertial Navigation (AAAI 2025 · similarity 0.80)
- Loose Inertial Poser: Motion Capture with IMU-attached Loose-Wear Jacket (CVPR 2024 · similarity 0.80)
- UMotion: Uncertainty-driven Human Motion Estimation from Inertial and Ultra-wideband Units (CVPR 2025 · similarity 0.79)
- MODA: Motion-Drift Augmentation for Inertial Human Motion Analysis (CVPR 2025 · similarity 0.79)
- EqNIO: Subequivariant Neural Inertial Odometry (ICLR 2025 · similarity 0.79)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
