Research map / 3D vision / Human pose estimation
"BodySLAM: Joint Camera Localisation, Mapping, and Human Motion Tracking"
Dorian F. Henning, Tristan Laidlow, Stefan Leutenegger
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 the human motion model factor (ablation) | monocular bundle adjustment baseline · BodySLAM dataset (10 sequences) | root trajectory error (H-ATE): 48.3% average improvement | Improved |
| add the human motion model factor (ablation) | monocular bundle adjustment baseline · BodySLAM dataset (10 sequences) | joint trajectory error (J-ATE): 35.6% average improvement | Improved |
| add the human motion model factor (ablation) | monocular bundle adjustment baseline · BodySLAM dataset, seq E 1 (SE(3) alignment) | C-ATE (mm): 108.1 -> 44.3 | Improved |
| add the human motion model factor (ablation) | monocular bundle adjustment baseline · BodySLAM dataset (10 sequences) | camera scale error: 0.35 to 0.74 -> 0.93 to 1.03 | Improved |
| add the human motion model factor (ablation) | monocular bundle adjustment baseline · BodySLAM dataset seq M 3 (Sim(3) alignment) | C-ATE (mm): 128 -> 128.1 | No clear effect |
| add the human motion model factor (ablation) | monocular bundle adjustment baseline · BodySLAM dataset seq E 1 (Sim(3) alignment) | H-ATE (mm): 106.5 -> 56.0 | Improved |
| add structural landmarks to the joint optimisation (ablation) | BodySLAM · BodySLAM dataset (5 perturbed sequences) | mean C-ATE (mm, SE(3)): 158.662 -> 155.572 | Improved |
| perturb camera scale by factor 5.0 before joint optimisation (ablation) | BodySLAM · BodySLAM dataset scale recovery | scale deviation (%): 5.1 -> 26.5 | Hurt |
| perturb camera scale by factor 10 (1/s' = 0.1) before joint optimisation (ablation) | BodySLAM · BodySLAM dataset scale recovery | scale deviation (%): 5.1 -> 24.0 | Hurt |
| use small scale perturbation (1/s' from 0.5 to 2.0) (ablation) | BodySLAM · BodySLAM dataset scale recovery | scale deviation (%): 5.1 to 5.6 | No clear effect |
Closest papers by meaning
- HumanBA: Human-Aware Bundle Adjustment via Global Human-Camera Decoupling (CVPR 2026 · similarity 0.88)
- Optimizing Human Pose Estimation Through Focused Human and Joint Regions (AAAI 2025 · similarity 0.88)
- Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation From Monocular Video (CVPR 2022 · similarity 0.87)
- Trajectory Optimization for Physics-Based Reconstruction of 3D Human Pose From Monocular Video (CVPR 2022 · similarity 0.87)
- Motions as Queries: One-Stage Multi-Person Holistic Human Motion Capture (CVPR 2025 · similarity 0.87)
- HumanMM: Global Human Motion Recovery from Multi-shot Videos (CVPR 2025 · 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.
