Research map / Point clouds / Shape and surface completion
From Point Clouds to Mesh Using Regression
Lubor Ladicky, Olivier Saurer, SoHyeon Jeong, Fabio Maninchedda, Marc Pollefeys
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 |
|---|---|---|---|
| train regressor on only 4 handpicked models instead of 102 models (ablation) | Regression forest · ModelNet40 synthetic, no noise | projection error (|c|): 0.66|c| error with no noise | Mixed |
| add 2% outliers to the synthetic test point clouds (ablation) | Regression forest · ModelNet40 synthetic test | reconstruction performance: 2% | No clear effect |
| increase Gaussian noise from no noise to sigma = 2.0|c| (ablation) | Regression forest · ModelNet40 synthetic test | mean projection error (|c|): 0.754 -> 1.910 | Hurt |
| increase distance of samples from surface from 0-2|c| to 8-10|c| (ablation) | Regression forest · ModelNet40 synthetic, no noise | projection error (|c|): 0.440 -> 1.128 | Hurt |
| use 4 trees instead of 1 tree (ablation) | Regression forest · ModelNet40 synthetic test | performance: 4 trees of maximum depth 20 vs only one tree | No clear effect |
| skip forest evaluation in empty boxes using integral volumes (K = 8) | Adaptive evaluation · Real mobile-phone scans | forest evaluations: avoid 90% of forest evaluations | Improved |
| regress signed distance directly instead of projection difference (ablation) | Regression forest · Synthetic data | reconstruction quality | Hurt |
| regress absolute distance instead of projection difference (ablation) | Regression forest · Synthetic data | surface extraction | Hurt |
Closest papers by meaning
- A Survey and Benchmark of Automatic Surface Reconstruction From Point Clouds (TPAMI 2024 · similarity 0.90)
- Scalable Surface Reconstruction From Point Clouds With Extreme Scale and Density Diversity (CVPR 2017 · similarity 0.90)
- Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learning (AAAI 2024 · similarity 0.88)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (CVPR 2025 · similarity 0.88)
- GridFormer: Point-Grid Transformer for Surface Reconstruction (AAAI 2024 · similarity 0.88)
- High-Fidelity Lightweight Mesh Reconstruction from Point Clouds (CVPR 2025 · 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.
