Research map / 3D vision / Human pose estimation
PyMAF: 3D Human Pose and Shape Regression With Pyramidal Mesh Alignment Feedback Loop
Hongwen Zhang, Yating Tian, Xinchi Zhou, Wanli Ouyang, Yebin Liu, Limin Wang, Zhenan Sun
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 mesh-aligned features instead of grid features (no pyramid) (ablation) | PyMAF feedback loop · Human3.6M | MPJPE (mm): 80.5 -> 79.6 | Improved |
| use mesh-aligned instead of grid features with pyramid (ablation) | PyMAF · Human3.6M | MPJPE (mm): 79.7 -> 76.8 | Improved |
| use grid instead of global features (ablation) | feedback loop · Human3.6M | MPJPE (mm): 84.3 -> 80.5 | Improved |
| add feature pyramid and 3 regressors to mesh-aligned feedback (ablation) | PyMAF · Human3.6M | MPJPE (mm): 79.6 -> 76.8 | Improved |
| add auxiliary dense correspondence supervision (ablation) | PyMAF · Human3.6M | MPJPE (mm): 76.8 -> 75.1 | Improved |
| add auxiliary part segmentation supervision (ablation) | PyMAF · Human3.6M | MPJPE (mm): 76.8 -> 75.5 | Improved |
| use dense correspondences instead of part segmentation for auxiliary supervision (ablation) | PyMAF · Human3.6M | MPJPE (mm): 75.5 -> 75.1 | Improved |
| use feature maps instead of auxiliary predictions as regressor input (ablation) | PyMAF · Human3.6M | MPJPE (mm): 77.8 -> 75.1 | Improved |
| mask mesh-aligned features with foreground part segmentation (ablation) | PyMAF · Human3.6M | MPJPE (mm): 75.5 -> 77.6 | Hurt |
| add auxiliary supervision (ablation) | PyMAF · 3DPW | MPJPE (mm): 95.6 -> 92.8 | Improved |
| add auxiliary supervision (ablation) | PyMAF · Human3.6M | MPJPE (mm): 60.3 -> 57.7 | Improved |
| add auxiliary supervision (ablation) | PyMAF · 3DPW | PA-MPJPE (mm): 58.8 -> 58.9 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports PyMAF * as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- Towards Robust and Expressive Whole-body Human Pose and Shape Estimation (NeurIPS 2023 · better on 0 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SPIN — SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition (AAAI 2021) · better on 8 of 8 shared results
- VIBE — VIBE: Video Inference for Human Body Pose and Shape Estimation (CVPR 2020) · better on 2 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PyMAF-X: Towards Well-Aligned Full-Body Model Regression From Monocular Images (TPAMI 2023 · similarity 0.94)
- ADHMR: Aligning Diffusion-based Human Mesh Recovery via Direct Preference Optimization (ICML 2025 · similarity 0.86)
- TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation (CVPR 2024 · similarity 0.86)
- GenHMR: Generative Human Mesh Recovery (AAAI 2025 · similarity 0.86)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · similarity 0.85)
- CUPS: Improving Human Pose-Shape Estimators with Conformalized Deep Uncertainty (ICML 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.
