Research map / 3D vision / Human pose estimation
mRI: Multi-modal 3D Human Pose Estimation Dataset using mmWave, RGB-D, and Inertial Sensors
Sizhe An, Yin Li, Umit Ogras
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 |
|---|---|---|---|
| fuse all three modalities (mmWave+RGB+IMUs) instead of best single modality for action detection (ablation) | ActionFormer on estimated 3D poses · mRI S2 Protocol 1 | average mAP (%): 93.46 -> 94.88 | Improved |
| fuse all three modalities instead of best single modality (ablation) | ActionFormer · mRI S2 Protocol 2 | average mAP (%): 95.07 -> 95.83 | Improved |
| fuse all three modalities instead of IMUs alone (ablation) | ActionFormer · mRI S2 Protocol 1 | mAP at tIoU=0.95: 53.55 -> 60.62 | Improved |
| fuse IMUs with RGB instead of RGB alone (ablation) | ActionFormer · mRI S2 Protocol 2 | average mAP (%): 95.07 -> 94.80 | No clear effect |
| split by subjects (S2) instead of random split (S1) for IMU-based pose estimation | IMU HPE · mRI Protocol 1 | MPJPE (mm): 80.2 -> 147.4 | Hurt |
| split by subjects (S2) instead of random split (S1) for mmWave pose estimation | mmWave HPE · mRI Protocol 2 | MPJPE (mm): 125.1 -> 126.6 | No clear effect |
| restrict to fixed rehabilitation movements (Protocol 2) instead of all 12 movements | mmWave HPE · mRI S1 | MPJPE (mm): 163.3 -> 125.1 | Improved |
| restrict to fixed rehabilitation movements (Protocol 2) instead of all 12 movements | IMU HPE · mRI S1 | MPJPE (mm): 80.2 -> 40.9 | Improved |
Closest papers by meaning
- MVDoppler-Pose: Multi-Modal Multi-View mmWave Sensing for Long-Distance Self-Occluded Human Walking Pose Estimation (CVPR 2025 · similarity 0.86)
- FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions (CVPR 2024 · similarity 0.85)
- HiPoser: 3D Human Pose Estimation with Hierarchical Shared Learning at Parts-Level Using Inertial Measurement Units (AAAI 2025 · similarity 0.85)
- Dynamic Inertial Poser (DynaIP): Part-Based Motion Dynamics Learning for Enhanced Human Pose Estimation with Sparse Inertial Sensors (CVPR 2024 · similarity 0.84)
- UMotion: Uncertainty-driven Human Motion Estimation from Inertial and Ultra-wideband Units (CVPR 2025 · similarity 0.84)
- Multi-View 3D Human Pose Estimation with Weakly Synchronized Images (AAAI 2025 · similarity 0.84)
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.
