Research map / Point clouds / Shape and surface completion
Normal Integration via Inverse Plane Fitting With Minimum Point-to-Plane Distance
Xu Cao, Boxin Shi, Fumio Okura, Yasuyuki Matsushita
Check 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 |
|---|---|---|---|
| Minimize squared point-to-plane distances instead of vertical-distance residuals | Normal integration · Anisotropic Gaussian surface with noise and 1% outliers | RMSE: 5.30 / 3.97 -> 0.86 / 0.69 ×10−2 | Improved |
| Use four-point plane fitting on the DGP domain | Normal integration · Sphere with background padded | RMSE: 8.24 / 5.21 -> 2.79 / 0.64 ×10−1 | Improved |
| Minimize squared point-to-plane distances | Perspective normal integration · SPHERE | RMSE: 0.54 / 0.38 -> 0.35 / 0.20 ×10−2 | Improved |
| Use four-point plane fitting on the DGP domain | Perspective normal integration · BUNNY | RMSE: 3.77 / 1.49 -> 3.75 / 1.49 ×10−3 | No clear effect |
| Use four-point plane fitting on the DGP domain | Perspective normal integration · HARVEST | RMSE: 12.90 / 11.11 -> 12.91 / 11.11 ×1 | No clear effect |
| Allow independent local offsets between horizontal and vertical neighbor pairs (ablation) | Five-point plane-fitting formulation · Checkerboard artifact test | Reconstruction error | Hurt |
| Use central-difference discretization rather than shared plane constraints (ablation) | Normal integration · Checkerboard artifact test | Checkerboard artifact | Hurt |
| Use a higher-order Savitzky-Golay filter instead of central differences (ablation) | Zhu and Smith perspective integration · Perspective normal maps | Reconstruction quality | Hurt |
Closest papers by meaning
- Discontinuity-aware Normal Integration for Generic Central Camera Models (ICCV 2025 · similarity 0.91)
- Feature-Preserving Mesh Decimation for Normal Integration (CVPR 2025 · similarity 0.89)
- PMNI: Pose-free Multi-view Normal Integration for Reflective and Textureless Surface Reconstruction (CVPR 2025 · similarity 0.86)
- Discontinuity-preserving Normal Integration with Auxiliary Edges (CVPR 2024 · similarity 0.86)
- SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration (CVPR 2024 · similarity 0.84)
- RNb-NeuS: Reflectance and Normal-based Multi-View 3D Reconstruction (CVPR 2024 · similarity 0.84)
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.
