Research map / Point clouds / Shape and surface completion
Human as Points: Explicit Point-Based 3D Human Reconstruction From Single-View RGB Images
Yingzhi Tang, Qijian Zhang, Yebin Liu, Junhui Hou
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 |
|---|---|---|---|
| replace implicit methods with explicit point-based HaP | PIFu/ICON/ECON · Thuman3.0 | CD: 1.306 -> 0.860 | Improved |
| add SMPL rectification (HaP-refined SMPL) to ICON (ablation) | ICON · Thuman3.0 | CD: 1.306 -> 0.982 | Improved |
| add HaP-rectified SMPL to ECON (ablation) | ECON · Thuman3.0 | CD: 1.467 -> 0.970 | Improved |
| add SMPL rectification module (ablation) | HaP · Thuman3.0 | CD: 1.051 -> 0.860 | Improved |
| use point-to-face with keypoint regularization instead of P2F only (ablation) | SMPL rectification · Thuman2 | CD: 1.056 -> 0.818 | Improved |
| add 2D keypoint regularization on top of P2F (ablation) | SMPL rectification · Thuman2 | CD: 0.823 -> 0.818 | Improved |
| use diffusion PNet instead of VRCNet for generation (ablation) | HaP · Thuman3.0 | Normal: 9.352 -> 2.170 | Improved |
| add depth replacement (ablation) | HaP · Thuman3.0 | Normal: 7.082 -> 2.170 | Improved |
| add displacement refinement network PNet Theta2 (ablation) | HaP · Thuman3.0 | Normal: 7.437 -> 2.170 | Improved |
| use MIM instead of P3Depth for depth estimation (ablation) | HaP · Thuman3.0 | CD: 0.946 -> 0.860 | Improved |
| use PyMAF instead of PIXIE for SMPL estimation (ablation) | HaP · Thuman3.0 | CD: 0.883 -> 0.860 | Improved |
| use HaP-I (512 input) versus IntegratedPIFu (1024 input) | IntegratedPIFu · Thuman3.0 | CD: 1.148 -> 1.021 | Improved |
Closest papers by meaning
- End-to-End Recovery of Human Shape and Pose (CVPR 2018 · similarity 0.90)
- PointHPS: Cascaded 3D Human Pose and Shape Estimation from Point Clouds (IJCV 2026 · similarity 0.90)
- Skeleton-Aware 3D Human Shape Reconstruction From Point Clouds (ICCV 2019 · similarity 0.90)
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.88)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.88)
- Neural Localizer Fields for Continuous 3D Human Pose and Shape Estimation (NeurIPS 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.
