Research map / 3D vision / Human pose estimation
Group Inertial Poser: Multi-Person Pose and Global Translation from Sparse Inertial Sensors and Ultra-Wideband Ranging
Ying Xue, Jiaxi Jiang, Rayan Armani, Dominik Hollidt, Yi-Chi Liao, Christian Holz
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 initial position optimization (ablation) | GIP · InterHuman, initial position unknown | Trans-3m (cm): 99.50 -> 63.85 | Improved |
| add trajectory optimization using between-person distances (ablation) | GIP · InterHuman, initial position unknown | RMSE (cm): 40.00 -> 34.86 | Improved |
| add trajectory optimization (ablation) | GIP · InterHuman, initial position unknown | Trans-6m (cm): 115.32 -> 98.93 | Improved |
| use SSM (S4) modules instead of LSTM (ablation) | GIP pose estimator · InterHuman | Angle Err (deg): 12.88 -> 9.94 | Improved |
| use SSM (S4) modules instead of LSTM (ablation) | GIP pose estimator · InterHuman | Trans-3m (cm): 66.60 -> 63.85 | Improved |
| add between-person UWB distances with trajectory optimization versus UIP/PIP baselines | GIP · InterHuman | Dist Err-20s (cm): 81.44 -> 1.91 | Improved |
| use initial position optimizer instead of ground-truth initialization | GIP · GIP-DB Dist Err-20s | Dist Err-20s (cm): 20.69 -> 20.71 | No clear effect |
| fine-tune on GIP-DB real data | GIP · GIP-DB | Angle Err (deg): 23.34 -> 17.57 | Improved |
| add more people to the joint optimization (ablation) | GIP · EgoHumans four-person synthetic | Trans Error (m): 2.34 -> 1.15 | Improved |
Closest papers by meaning
- Ultra Diffusion Poser: Diffusion-Based Human Motion Tracking from Sparse Inertial Sensors and Ranging-based Between-sensor Distances (CVPR 2026 · similarity 0.91)
- UMotion: Uncertainty-driven Human Motion Estimation from Inertial and Ultra-wideband Units (CVPR 2025 · similarity 0.90)
- Dynamic Inertial Poser (DynaIP): Part-Based Motion Dynamics Learning for Enhanced Human Pose Estimation with Sparse Inertial Sensors (CVPR 2024 · similarity 0.90)
- Motions as Queries: One-Stage Multi-Person Holistic Human Motion Capture (CVPR 2025 · similarity 0.88)
- I'M HOI: Inertia-aware Monocular Capture of 3D Human-Object Interactions (CVPR 2024 · similarity 0.87)
- HiPoser: 3D Human Pose Estimation with Hierarchical Shared Learning at Parts-Level Using Inertial Measurement Units (AAAI 2025 · similarity 0.87)
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.
