Research map / Point clouds / Shape and surface completion
Reconstructing Thin Structures of Manifold Surfaces by Integrating Spatial Curves
Shiwei Li, Yao Yao, Tian Fang, Long Quan
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 |
|---|---|---|---|
| integrate reconstructed spatial curves with MVS points for surface reconstruction | Ours-Curve · chair | F1: 95.21 -> 95.65 | Improved |
| integrate reconstructed spatial curves with MVS points for surface reconstruction | Ours-Curve · chair2 | F1: 74.84 -> 76.45 | Improved |
| integrate reconstructed spatial curves with MVS points for surface reconstruction | Ours-Curve · fence | F1: 76.78 -> 84.09 | Improved |
| integrate reconstructed spatial curves with MVS points for surface reconstruction | Ours-Curve · holder | F1: 93.74 -> 93.99 | Improved |
| integrate reconstructed spatial curves with MVS points for surface reconstruction | Ours-Curve · lamp | F1: 77.71 -> 79.41 | Improved |
| integrate reconstructed spatial curves with MVS points for surface reconstruction | Ours-Curve · pylon | F1: 77.61 -> 82.01 | Improved |
| optimize and extend 3D curves instead of using initial 3D line segments (ablation) | Ours-Curve · six synthetic datasets | F1: 91.70 -> 95.65; 72.89 -> 76.45; 79.02 -> 84.09; 89.21 -> 93.99; 74.27 -> 79.41; 81.07 -> 82.01 | Improved |
| use curve-conformed Delaunay refinement instead of uniform point sampling in curve confidence regions (ablation) | Ours-Curve · six synthetic datasets | F1: 91.87 -> 95.65; 75.25 -> 76.45; 81.68 -> 84.09; 90.63 -> 93.99; 77.42 -> 79.41; 80.51 -> 82.01 | Improved |
| integrate spatial curves rather than relying on point-based reconstruction | Ours-Curve vs Vu · six synthetic datasets | accuracy | Mixed |
Closest papers by meaning
- Structure-Aware Surface Reconstruction via Primitive Assembly (ICCV 2023 · similarity 0.88)
- Surface Reconstruction From Point Clouds: A Survey and a Benchmark (TPAMI 2024 · similarity 0.88)
- From Multiview Image Curves to 3D Drawings (ECCV 2016 · similarity 0.87)
- High-Fidelity Lightweight Mesh Reconstruction from Point Clouds (CVPR 2025 · similarity 0.87)
- Learning Delaunay Surface Elements for Mesh Reconstruction (CVPR 2021 · similarity 0.87)
- CrossSDF: 3D Reconstruction of Thin Structures From Cross-Sections (CVPR 2025 · 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.
