Research map / 3D vision / Human pose estimation
PointHPS: Cascaded 3D Human Pose and Shape Estimation from Point Clouds
Zhongang Cai, Liang Pan, Wanqi Yin, Chen Wei, Fangzhou Hong, Mingyuan Zhang, Atsushi Yamashita, Chen Change Loy, Lei Yang, Ziwei Liu
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 |
|---|---|---|---|
| stack a second stage without CFF or IFE (ablation) | one-stage PointHPS · HuMMan-Point | PVE: 111.2 -> 113.3 | Hurt |
| add CFF to the cascade (ablation) | naive cascade · HuMMan-Point | PVE: 113.3 -> 85.6 | Improved |
| add IFE to the cascade without CFF (ablation) | naive cascade · HuMMan-Point | PVE: 113.3 -> 104.8 | Improved |
| add IFE on top of cascade + CFF (ablation) | PointHPS · HuMMan-Point | PVE: 85.6 -> 81.9 | Improved |
| add upsampling-phase features in CFF (ablation) | cascade + IFE · HuMMan-Point | PA-MPJPE: 67.1 -> 62.9 | Improved |
| add downsampling-phase features in CFF (ablation) | CFF Up · HuMMan-Point | PVE: 104.1 -> 93.1 | Improved |
| add adaptive feature fusion in CFF (ablation) | CFF Up+Down · HuMMan-Point | PVE: 93.1 -> 81.9 | Improved |
| increase stages from 1 to 2 (ablation) | PointHPS · HuMMan-Point | PVE: 111.2 -> 81.9 | Improved |
| increase stages from 2 to 3 (ablation) | PointHPS · HuMMan-Point | PVE: 81.9 -> 79.1 | Improved |
| use PointHPS instead of VoteHMR | PointHPS · HuMMan-Point | PVE: 97.6 -> 81.9 | Improved |
| use PointHPS instead of VoteHMR | PointHPS · GTA-Human-Point | PVE: 114.4 -> 107.0 | Improved |
| use PointHPS instead of VoteHMR | PointHPS · SURREAL | PVE: 20.2 -> 19.6 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- 3DCODED — 3D-CODED: 3D Correspondences by Deep Deformation (ECCV 2018) · better on 2 of 2 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
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.89)
- Pre-training a Density-Aware Pose Transformer for Robust LiDAR-based 3D Human Pose Estimation (AAAI 2025 · similarity 0.88)
- LiveHPS: LiDAR-based Scene-level Human Pose and Shape Estimation in Free Environment (CVPR 2024 · similarity 0.88)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 2024 · similarity 0.88)
- Neighborhood-Enhanced 3D Human Pose Estimation with Monocular LiDAR in Long-Range Outdoor Scenes (AAAI 2024 · similarity 0.87)
- FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions (CVPR 2024 · similarity 0.87)
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.
