Research map / 3D vision / Human pose estimation
From Image to Stability: Learning Dynamics from Human Pose
Jesse Scott, Bharadwaj Ravichandran, Christopher Funk, Robert T. Collins, Yanxi Liu
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 |
|---|---|---|---|
| use 3D pose input instead of 2D (ablation) | PressNet-Simple · Taiji foot pressure (LOSO, mean over 10 subjects) | MAE (kPa): 8.95 -> 8.48 | Improved |
| use 3D pose input instead of 2D (ablation) | PressNet-Simple · Taiji foot pressure (LOSO, mean over 10 subjects) | KLD: 3.01 -> 2.25 | Improved |
| add BioPose biomechanical correction to 3D joints (ablation) | PressNet-Simple 3D · Taiji foot pressure (LOSO, mean over 10 subjects) | MAE (kPa): 8.48 -> 8.50 | No clear effect |
| add BioPose biomechanical correction to 3D joints (ablation) | PressNet-Simple 3D · Taiji foot pressure (LOSO, mean over 10 subjects) | KLD: 2.25 -> 2.68 | Hurt |
| apply BioPose correction to OpenPose 3D joints | OpenPose triangulation · 3D joint error vs motion capture | L2 error (mm): 53.6 -> 33.2 | Improved |
| use PressNet with KL loss instead of PressNet-Simple (both 3D BioPose) | PressNet-Simple 3D + BioPose · Taiji foot pressure (LOSO, mean over 10 subjects) | KLD: 2.68 -> 1.11 | Improved |
| use PressNet-Simple 2D instead of KNN baseline (K=5) | KNN · Taiji foot pressure (LOSO, mean over 10 subjects) | MAE (kPa): 10.06 -> 8.95 | Improved |
| use PressNet-Simple 2D instead of KNN baseline (K=5) | KNN · Taiji foot pressure (LOSO, mean over 10 subjects) | KLD: 14.35 -> 3.01 | Improved |
| test on unseen camera view 2 after training on view 1 | PressNet-Simple 2D · Taiji cross-view | MAE (kPa): 8.95 -> 9.23 | No clear effect |
| test on unseen camera view 2 after training on view 1 | PressNet-Simple 2D · Taiji cross-view | KLD: 3.01 -> 2.80 | No clear effect |
Closest papers by meaning
- Bodies at Rest: 3D Human Pose and Shape Estimation From a Pressure Image Using Synthetic Data (CVPR 2020 · similarity 0.87)
- MotionPRO: Exploring the Role of Pressure in Human MoCap and Beyond (CVPR 2025 · similarity 0.86)
- Pressure2Motion: Hierarchical Human Motion Reconstruction from Ground Pressure with Text Guidance (CVPR 2026 · similarity 0.85)
- MMVP: A Multimodal MoCap Dataset with Vision and Pressure Sensors (CVPR 2024 · similarity 0.85)
- BodyMAP - Jointly Predicting Body Mesh and 3D Applied Pressure Map for People in Bed (CVPR 2024 · similarity 0.84)
- EgoPressure: A Dataset for Hand Pressure and Pose Estimation in Egocentric Vision (CVPR 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.
