Research map / 3D vision / Human pose estimation
Dynamics-regulated kinematic policy for egocentric pose estimation
Zhengyi Luo, Ryo Hachiuma, Ye Yuan, Kris M. Kitani
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 dynamics-regulated training with simulated states to supervised training (ablation) | Kin_poly kinematic policy · Real-world dataset | S_inter: 73.2% -> 80.9% | Improved |
| Add RL updates to dynamics-regulated training (ablation) | Kin_poly kinematic policy · Real-world dataset | S_inter: 80.9% -> 93.4% | Improved |
| Use VIO camera pose instead of optical-flow image features (ablation) | Kin_poly kinematic policy · Real-world dataset | S_inter: 54.1% -> 93.4% | Improved |
| Use VIO camera pose instead of optical-flow image features (ablation) | Kin_poly kinematic policy · Real-world dataset | E_cam: 1.129 -> 0.475 | Improved |
| Use dynamics-regulated training rather than supervised-only kinematics | Kin_poly · MoCap dataset | S_inter | Improved |
| Use dynamics-regulated training rather than supervised-only kinematics | Kin_poly · MoCap dataset | E_mpjpe: 33.149 -> 39.575 | Hurt |
| Use dynamics-regulated training rather than supervised-only kinematics | Kin_poly · MoCap dataset | PT: 10.076 -> 0.679 | Improved |
| Use Kin_poly supervised learning instead of PoseReg | Kin_poly kinematic policy · MoCap dataset | E_root: 0.857 -> 0.176 | Improved |
| Use Kin_poly dynamics-regulated training instead of EgoPose | Egocentric pose estimation · Real-world dataset | S_inter: 9.3% -> 93.4% | Improved |
Closest papers by meaning
- 3D Ego-Pose Estimation via Imitation Learning (ECCV 2018 · similarity 0.89)
- EgoPoseFormer v2: Accurate Egocentric Human Motion Estimation for AR/VR (CVPR 2026 · similarity 0.89)
- FRAME: Floor-aligned Representation for Avatar Motion from Egocentric Video (CVPR 2025 · similarity 0.89)
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.88)
- Estimating Body and Hand Motion in an Ego-sensed World (CVPR 2025 · similarity 0.87)
- EgoGen: An Egocentric Synthetic Data Generator (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.
