Research map / 3D vision / Human pose estimation
SMAP: Single-Shot Multi-Person Absolute 3D Pose Estimation
Jianan Zhen, Qi Fang, Jiaming Sun, Wentao Liu, Wei Jiang, Hujun Bao, Xiaowei Zhou
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 depth normalization by FoV size (ablation) | SMAP · MuPoTS-3D | PCK_abs: 8.7 -> 38.7 | Improved |
| add depth normalization by FoV size (ablation) | SMAP · MuPoTS-3D | PCK_root: 5.7 -> 45.5 | Improved |
| add multi-scale supervision (ablation) | SMAP · MuPoTS-3D | PCK_rel: 75.4 -> 80.5 | Improved |
| add multi-scale supervision (ablation) | SMAP · MuPoTS-3D | PCK_abs: 36.2 -> 38.7 | Improved |
| add RefineNet (ablation) | SMAP · MuPoTS-3D | PCK_rel: 70.9 -> 80.5 | Improved |
| add RefineNet (ablation) | SMAP · MuPoTS-3D | PCK_abs: 34.7 -> 38.7 | Improved |
| add RefineNet (ablation) | SMAP · CMU Panoptic | MPJPE (mm): 73.1 -> 61.8 | Improved |
| use larger backbone instead of smaller one (ablation) | SMAP · MuPoTS-3D | PCK_abs: 35.1 -> 38.7 | Improved |
| replace 2D part association with depth-aware part association (ablation) | SMAP · MuPoTS-3D | PCK_abs: 38.6 -> 38.7 | Improved |
| replace 2D part association with depth-aware part association (ablation) | SMAP · CMU Panoptic | Recall: 94.3 -> 96.4 | Improved |
| add 2D keypoint/PAF branches to root-depth-only network (ablation) | SMAP · MuPoTS-3D | PCK_root: 29.9 -> 43.6 | Improved |
| add part relative-depth branch to root depth + 2D branches (ablation) | SMAP · MuPoTS-3D | PCK_root: 43.6 -> 45.5 | Improved |
Closest papers by meaning
- Dual Networks Based 3D Multi-Person Pose Estimation From Monocular Video (TPAMI 2022 · similarity 0.90)
- Shape-Aware Multi-Person Pose Estimation From Multi-View Images (ICCV 2021 · similarity 0.89)
- Single-Stage Is Enough: Multi-Person Absolute 3D Pose Estimation (CVPR 2022 · similarity 0.89)
- Monocular 3D Multi-Person Pose Estimation by Integrating Top-Down and Bottom-Up Networks (CVPR 2021 · similarity 0.89)
- Coherent Reconstruction of Multiple Humans From a Single Image (CVPR 2020 · similarity 0.88)
- Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation (AAAI 2020 · 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.
