Research map / 3D vision / Human pose estimation
Recovering Accurate 3D Human Pose in The Wild Using IMUs and a Moving Camera
Timo von Marcard, Roberto Henschel, Michael J. Black, Bodo Rosenhahn, Gerard Pons-Moll
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 video with the IMU-only tracker via joint optimization (VIP vs IT) (ablation) | inertial tracker · TotalCapture | MPJPE (mm): 55.0 -> 26.0 | Improved |
| fuse video with the IMU-only tracker (VIP vs IT) (ablation) | inertial tracker · TotalCapture | MPJAE (degrees): 16.9 -> 12.1 | Improved |
| add optimized heading angles and initial pose to IT (IT vs VIP-IT) (ablation) | inertial tracker · TotalCapture | MPJPE (mm): 55.0 -> 28.2 | Improved |
| add image-based optimization on top of VIP-IT (VIP vs VIP-IT) (ablation) | VIP-IT · TotalCapture | MPJPE (mm): 28.2 -> 26.0 | Improved |
| use ground-truth 2D poses instead of detected 2D poses (ablation) | VIP · TotalCapture | MPJPE (mm): 26.0 -> 15.1 | Improved |
| use ground-truth camera pose instead of estimated camera pose (ablation) | VIP · TotalCapture | MPJPE (mm): 26.0 -> 25.3 | No clear effect |
| use 13 IMUs instead of 6 IMUs (ablation) | VIP · TotalCapture | MPJPE (mm): 39.6 -> 26.0 | Improved |
| use 13 IMUs instead of 6 IMUs (ablation) | VIP · TotalCapture | MPJAE (degrees): 15.3 -> 12.1 | Improved |
| replace the TotalCapture baseline [39] with single-camera VIP | pose tracking · TotalCapture | MPJPE (mm): 70.0 -> 26.0 | Improved |
| replace Gilbert et al. [16] (8 cameras) with single-camera VIP | pose tracking · TotalCapture | MPJPE (mm): 62 -> 26.0 | Improved |
Closest papers by meaning
- Fusing Visual and Inertial Sensors with Semantics for 3D Human Pose Estimation (IJCV 2018 · similarity 0.90)
- SPEC: Seeing People in the Wild With an Estimated Camera (ICCV 2021 · similarity 0.90)
- Human POSEitioning System (HPS): 3D Human Pose Estimation and Self-Localization in Large Scenes From Body-Mounted Sensors (CVPR 2021 · similarity 0.88)
- Exploiting Temporal Context for 3D Human Pose Estimation in the Wild (CVPR 2019 · similarity 0.87)
- HumanMM: Global Human Motion Recovery from Multi-shot Videos (CVPR 2025 · similarity 0.86)
- FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions (CVPR 2024 · similarity 0.86)
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.
