Research map / 3D vision / Human pose estimation
Probabilistic 3D Human Shape and Pose Estimation From Multiple Unconstrained Images in the Wild
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 |
|---|---|---|---|
| probabilistically combine five image shape distributions instead of using a single image (ablation) | Ours · SSP-3D | PVE-T-SC (mm): 15.2 -> 13.3 | Helped |
| probabilistically combine five image shape distributions instead of taking their simple mean (ablation) | Ours · SSP-3D | PVE-T-SC (mm): 13.6 -> 13.3 | Helped |
| increase the maximum image group size from one to five (ablation) | Ours + PC · SSP-3D | PVE-T-SC (mm): 15.2 -> 13.3 | Helped |
| use complementary-view pairs rather than redundant-view pairs (ablation) | Ours + PC · Synthetic inputs | PVE-T-SC (mm): 13.6 -> 13.5 | Helped |
| use probabilistic combination rather than simple averaging (ablation) | Ours · Synthetic clean and corrupted inputs | PVE-T-SC (mm): 13.1 -> 13.0; 13.3 -> 12.8 | Helped |
| use probabilistic combination rather than simple averaging (ablation) | Ours · Private tape-measure dataset | Chest RMSE (cm): 3.4 -> 3.1 | Helped |
| use probabilistic combination rather than simple averaging (ablation) | Ours · Private tape-measure dataset | Biceps RMSE (cm): 4.9 -> 5.0 | Hurt |
| increase the maximum image group size from one to five (ablation) | Ours + PC · SSP-3D | PVE-T-SC (mm): 15.2 -> 13.9 -> 13.6 -> 13.5 -> 13.3 | Helped |
| train with increased occlusion and cropping augmentations (ablation) | Ours · 3DPW | MPJPE-PA (mm) | Helped |
| set heatmaps for invisible joints to zero and supervise their visibility (ablation) | Ours · 3DPW | pose uncertainty | Helped |
| use predicted proxy inputs rather than ground-truth proxy inputs (ablation) | Ours · SSP-3D | PVE-T-SC (mm): 14.4 -> 15.2 | Hurt |
| use predicted proxy inputs rather than synthetic ground-truth proxy inputs (ablation) | Ours · 3DPW | MPJPE-PA (mm): 45.7 -> 61.0 | Hurt |
Closest papers by meaning
- Learning to Estimate 3D Human Pose and Shape From a Single Color Image (CVPR 2018 · similarity 0.92)
- Accurate 3D Body Shape Regression Using Metric and Semantic Attributes (CVPR 2022 · similarity 0.90)
- MUC: Mixture of Uncalibrated Cameras for Robust 3D Human Body Reconstruction (AAAI 2025 · similarity 0.88)
- CUPS: Improving Human Pose-Shape Estimators with Conformalized Deep Uncertainty (ICML 2025 · similarity 0.88)
- ProPose: Probabilistic 3D Human Pose Estimation with Instance-Level Distribution and Normalizing Flow (AAAI 2025 · similarity 0.88)
- Self-Supervised 3D Human Mesh Recovery from a Single Image with Uncertainty-Aware Learning (AAAI 2024 · 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.