Research map / 3D vision / Human pose estimation
PoseAug: A Differentiable Pose Augmentation Framework for 3D Human Pose Estimation
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 |
|---|---|---|---|
| Add PoseAug to VPose (1-frame) | VPose (1-frame) · H36M | MPJPE: 52.7 -> 50.2 | Helped |
| Add PoseAug to VPose (1-frame) | VPose (1-frame) · 3DHP cross-scenario | MPJPE: 86.6 -> 73.0 | Helped |
| Add PoseAug to each tested estimator | SemGCN, SimpleBaseline, ST-GCN, and VPose · 3DPW test | PA-MPJPE: 102.0 -> 82.2; 89.4 -> 78.1; 98.0 -> 73.2; 94.6 -> 81.6 | Helped |
| Add extra MPII 2D poses for training the 2D discriminator | PoseAug · 3DHP cross-scenario | MPJPE: 73.0 -> 71.1 | Helped |
| Add bone-angle augmentation (ablation) | PoseAug augmentor · 3DHP | MPJPE: 86.6 -> 85.2 | Helped |
| Add rigid transformation augmentation (ablation) | PoseAug augmentor · 3DHP | MPJPE: 86.6 -> 75.9 | Helped |
| Add bone-length augmentation to BA+RT (ablation) | PoseAug augmentor · 3DHP | MPJPE: 75.5 -> 73.5 | Helped |
| Add estimator error feedback to BA+RT+BL (ablation) | PoseAug augmentor · 3DHP | MPJPE: 73.5 -> 73.0 | Helped |
| Use the 2D discriminator (ablation) | PoseAug · 3DHP | MPJPE: 86.6 -> 80.8 | Helped |
| Use the 3D discriminator (ablation) | PoseAug · 3DHP | MPJPE: 86.6 -> 79.6 | Helped |
| Replace KCS with part-aware KCS (ablation) | PoseAug discriminator · 3DHP | MPJPE: 77.7 -> 73.0 | Helped |
| Replace KCS with part-aware KCS (ablation) | PoseAug discriminator · 3DPW | PA-MPJPE: 88.4 -> 81.6 | Helped |
Closest papers by meaning
- A Dual-Augmentor Framework for Domain Generalization in 3D Human Pose Estimation (CVPR 2024 · similarity 0.90)
- Toward Approaches to Scalability in 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.88)
- ProPose: Probabilistic 3D Human Pose Estimation with Instance-Level Distribution and Normalizing Flow (AAAI 2025 · similarity 0.88)
- ManiPose: Manifold-Constrained Multi-Hypothesis 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.88)
- PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization (CVPR 2024 · similarity 0.88)
- PoseBH: Prototypical Multi-Dataset Training Beyond Human Pose Estimation (CVPR 2025 · similarity 0.88)
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.