Research map / 3D vision / Human pose estimation
ActiveMoCap: Optimized Viewpoint Selection for Active Human Motion Capture
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| choose camera viewpoints by minimizing predicted pose uncertainty | ActiveMoCap · Teleportation, total | 3D pose error: 0.195±0.07 -> 0.144±0.35; 0.216±0.06 -> 0.144±0.35 | Helped |
| choose camera viewpoints by minimizing predicted pose uncertainty | ActiveMoCap · MPI-INF-3DHP, teleportation (networks) | 3D pose error: 0.28±0.03 -> 0.21±0.0008; 0.29±0.03 -> 0.21±0.0008 | Helped |
| choose camera viewpoints by minimizing predicted pose uncertainty | ActiveMoCap · MPI-INF-3DHP, teleportation (noisy ground truth) | 3D pose error: 0.286±0.027 -> 0.145±0.006; 0.265±0.010 -> 0.145±0.006 | Helped |
| use uncertainty-based active viewpoint selection rather than constant rotation | ActiveMoCap · Drone flight, total | 3D pose error: 0.30±0.08 -> 0.31±0.10 | No clear effect |
| use uncertainty-based active viewpoint selection rather than random movement | ActiveMoCap · Drone flight, total | 3D pose error: 0.62±0.15 -> 0.31±0.10 | Helped |
| add a drone flight model to predict future camera positions (ablation) | ActiveMoCap · CMU-Dribble, CMU-Sitting, CMU-Dinosour; drone flight | 3D pose error: 0.65±0.09 -> 0.28±0.006; 0.48±0.09 -> 0.15±0.007; 0.22±0.07 -> 0.12±0.02 | Helped |
| add a drone flight model to predict future camera positions (ablation) | ActiveMoCap · Drone flight, total (ablation study) | 3D pose error: 0.45±0.08 -> 0.18±0.01 | Helped |
Closest papers by meaning
- Domes to Drones: Self-Supervised Active Triangulation for 3D Human Pose Reconstruction (NeurIPS 2019 · similarity 0.90)
- Deep Reinforcement Learning for Active Human Pose Estimation (AAAI 2020 · similarity 0.89)
- Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation From Monocular Video (CVPR 2022 · similarity 0.87)
- Learning Monocular 3D Human Pose Estimation From Multi-View Images (CVPR 2018 · similarity 0.87)
- FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions (CVPR 2024 · similarity 0.87)
- Motions as Queries: One-Stage Multi-Person Holistic Human Motion Capture (CVPR 2025 · similarity 0.85)
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.