Research map / Point clouds / Shape and surface completion
CircNet: Meshing 3D Point Clouds with Circumcenter Detection
Huan Lei, Ruitao Leng, Liang Zheng, Hongdong Li
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 candidate-triangle enumeration with single-shot circumcenter detection | CircNet · ABC test | total runtime/seconds: 37.683 -> 3.316 [0.996] | Improved |
| replace candidate-triangle enumeration with single-shot circumcenter detection | CircNet · MGN | CD1 (×10²) ↓: 0.270 -> 0.269 | Improved |
| use circumcenter detection and duality to recover triangles | CircNet · FAUST | F1 ↑: 0.995 -> 0.993 | Hurt |
| use circumcenter detection and duality to recover triangles | CircNet · ABC test | F1 ↑: 0.949 -> 0.950 | Improved |
| reduce neighborhood size K from 50 to 25 (ablation) | CircNet · ABC hyperparameter study | CD1 (×10²) ↓: 0.284 -> 0.285 | Hurt |
| reduce predictions per anchor cell s from 2 to 1 (ablation) | CircNet · ABC hyperparameter study | CD2 (×10⁵) ↓: 0.545 -> 0.636 | Hurt |
| increase neighborhood size K from 50 to 100 (ablation) | CircNet · ABC hyperparameter study | CD1 (×10²) ↓: 0.284 -> 0.284 | No clear effect |
| increase neighborhood size K from 50 to 200 (ablation) | CircNet · ABC hyperparameter study | CD1 (×10²) ↓: 0.284 -> 0.284 | No clear effect |
| increase Gaussian noise level | CircNet · ABC robustness | F1 ↑: 0.908 -> 0.793 -> 0.689 | Hurt |
Later papers that compared with this method
1 later accepted paper reports CircNet as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- OffsetOPT: Explicit Surface Reconstruction without Normals (CVPR 2025 · better on 1 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PointTriNet — PointTriNet: Learned Triangulation of 3D Point Sets (ECCV 2020) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PointTriNet: Learned Triangulation of 3D Point Sets (ECCV 2020 · similarity 0.89)
- Meshing Point Clouds with Predicted Intrinsic-Extrinsic Ratio Guidance (ECCV 2020 · similarity 0.89)
- Surface Reconstruction From Point Clouds: A Survey and a Benchmark (TPAMI 2024 · similarity 0.89)
- EdgeMovingNet: Edge-preserving Point Cloud Reconstruction via Joint Geometry Features (CVPR 2025 · similarity 0.88)
- LAM3D: Large Image-Point Clouds Alignment Model for 3D Reconstruction from Single Image (NeurIPS 2024 · similarity 0.87)
- 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.
