Research map / Point clouds / Shape and surface completion
SHS-Net: Learning Signed Hyper Surfaces for Oriented Normal Estimation of Point Clouds
Read the paperCheck 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 |
|---|---|---|---|
| use SHS-Net instead of PCA+MST | oriented normal estimation · PCPNet | oriented normal RMSE (Average): 28.52 -> 19.79 | Helped |
| use SHS-Net instead of PCA+MST | oriented normal estimation · FamousShape | oriented normal RMSE (Average): 40.48 -> 32.79 | Helped |
| use SHS-Net instead of HSurf-Net | unoriented normal estimation · PCPNet | unoriented normal RMSE (Average): 10.11 -> 9.96 | Helped |
| use SHS-Net instead of HSurf-Net | unoriented normal estimation · FamousShape | unoriented normal RMSE (Average): 17.70 -> 17.61 | Helped |
| add patch encoding (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 45.80 -> 19.79 | Helped |
| add shape encoding (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 76.37 -> 19.79 | Helped |
| add distance-based weight w (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 22.00 -> 19.79 | Helped |
| use attention-weighted normal prediction module H instead of simple MLP layers (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 23.11 -> 19.79 | Helped |
| separately constrain unoriented normal and sign rather than directly regress oriented normals (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 33.29 -> 19.79 | Helped |
| combine density-gradient and random-sample terms in shape point sampling (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 24.07 -> 19.79; 21.54 -> 19.79 | Helped |
| use the full random-sample ratio ζ=1/1.5 instead of ζ=1/2 or ζ=1/3 (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 23.67 -> 19.79; 23.05 -> 19.79 | Helped |
| use NP=1200 instead of NP=1100 or NP=1300 (ablation) | SHS-Net · PCPNet | oriented normal RMSE (Average): 21.69 -> 19.79; 21.16 -> 19.79 | Helped |
Closest papers by meaning
- HSurf-Net: Normal Estimation for 3D Point Clouds by Learning Hyper Surfaces (NeurIPS 2022 · similarity 0.93)
- Consistent Normal Orientation for 3D Point Clouds via Least Squares on Delaunay Graph (CVPR 2025 · similarity 0.86)
- An Explicit Frame Construction for Normalizing 3D Point Clouds (ICML 2024 · similarity 0.86)
- Learning Bijective Surface Parameterization for Inferring Signed Distance Functions from Sparse Point Clouds with Grid Deformation (CVPR 2025 · similarity 0.86)
- OffsetOPT: Explicit Surface Reconstruction without Normals (CVPR 2025 · similarity 0.86)
- CMG-Net: Robust Normal Estimation for Point Clouds via Chamfer Normal Distance and Multi-Scale Geometry (AAAI 2024 · similarity 0.86)
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.