Research map / 3D vision / Human pose estimation
Single-Stage Multi-Person Pose Machines
Xuecheng Nie, Jiashi Feng, Jianfeng Zhang, Shuicheng Yan
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 joint hierarchies (hierarchical SPR instead of plain SPR) (ablation) | SPM · MPII validation | mAP total (%): 75.3 -> 77.7 | Improved |
| add joint hierarchies (ablation) | SPM · MPII validation (wrists) | mAP (%): 65.2 -> 69.4 | Improved |
| add joint hierarchies (ablation) | SPM · MPII validation (ankles) | mAP (%): 60.3 -> 63.9 | Improved |
| add joint hierarchies (ablation) | SPM · MPII validation | time per image (s): 0.058 -> 0.058 | No clear effect |
| increase neighborhood size tau from 1 to 7 (ablation) | SPM · MPII validation | mAP | Improved |
| increase neighborhood size tau from 7 to 10 (ablation) | SPM · MPII validation | mAP | No clear effect |
| increase neighborhood size tau above 10 (ablation) | SPM · MPII validation | mAP | Hurt |
| single-stage SPM instead of bottom-up Newell and Deng | pose estimation · MPII full test | time per image (s): 0.25 -> 0.058 | Improved |
| single-stage SPM instead of Fieraru et al. | pose estimation · MPII full test | mAP total (%): 78.0 -> 78.5 | Improved |
| single-stage SPM instead of PersonLab | pose estimation · MSCOCO test-dev | AP: 0.687 -> 0.669 | Hurt |
| single-stage SPM instead of PersonLab | pose estimation · MSCOCO test-dev | time per image (s): 0.464 -> 0.058 | Improved |
| single-stage SPM instead of Xia et al. | pose estimation · extended PASCAL-Person-Part | mAP total (%): 39.2 -> 46.1 | Improved |
Closest papers by meaning
- DeeperCut: A Deeper, Stronger, and Faster Multi-person Pose Estimation Model (ECCV 2016 · similarity 0.90)
- End-to-End Multi-Person Pose Estimation With Transformers (CVPR 2022 · similarity 0.90)
- Dual Networks Based 3D Multi-Person Pose Estimation From Monocular Video (TPAMI 2022 · similarity 0.90)
- Direct Multi-view Multi-person 3D Pose Estimation (NeurIPS 2021 · similarity 0.89)
- MV-SSM: Multi-View State Space Modeling for 3D Human Pose Estimation (CVPR 2025 · similarity 0.88)
- SelfPose3d: Self-Supervised Multi-Person Multi-View 3d Pose Estimation (CVPR 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.
