Research map / Point clouds / Shape and surface completion
Delaunay Canopy: Building Wireframe Reconstruction from Airborne LiDAR Point Clouds via Delaunay Graph
Donghyun Kim, Chanyoung Kim, YoungJoong Kwon, Seong Jae Hwang
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 |
|---|---|---|---|
| add corner score sampling instead of FPS (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.257 -> 0.239 | Improved |
| add prior-based query scaling (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.257 -> 0.242 | Improved |
| add both corner score sampling and query scaling (ablation) | Delaunay Canopy · Building3D Tallinn city | EF1: 76.3 -> 80.0 | Improved |
| add query scaling on top of corner score sampling (ablation) | Delaunay Canopy · Building3D Tallinn city | CF1: 86.7 -> 88.3 | Improved |
| use own wire selection module instead of Point2Roof's (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.248 -> 0.239 | Improved |
| add Delaunay-derived prior scaling to wire module (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.239 -> 0.232 | Improved |
| increase sampled points K from 150 to 200 (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.232 -> 0.260 | Hurt |
| decrease sampled points K from 150 to 100 (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.232 -> 0.253 | Hurt |
| apply query scaling to 6 layers instead of first 3 (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.232 -> 0.255 | Hurt |
| apply query scaling to first 2 layers instead of 3 (ablation) | Delaunay Canopy · Building3D Tallinn city | WED: 0.230 vs 0.232 | No clear effect |
| use Delaunay Canopy instead of BWFormer on perturbed data | BWFormer · Building3D Tallinn city (perturbed sparse/noisy test) | WED: 0.339 -> 0.301 | Improved |
| use Delaunay Canopy instead of BWFormer | BWFormer · Building3D entry-level | CF1: 87.3 -> 90.2 | Improved |
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.
- BWFormer — BWFormer: Building Wireframe Reconstruction from Airborne LiDAR Point Cloud with Transformer (CVPR 2025) · better on 9 of 9 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
- BWFormer: Building Wireframe Reconstruction from Airborne LiDAR Point Cloud with Transformer (CVPR 2025 · similarity 0.92)
- PBWR: Parametric-Building-Wireframe Reconstruction from Aerial LiDAR Point Clouds (CVPR 2024 · similarity 0.91)
- EdgeDiff: Edge-aware Diffusion Network for Building Reconstruction from Point Clouds (CVPR 2025 · similarity 0.85)
- High-Fidelity Lightweight Mesh Reconstruction from Point Clouds (CVPR 2025 · similarity 0.85)
- LiDAR-Net: A Real-scanned 3D Point Cloud Dataset for Indoor Scenes (CVPR 2024 · similarity 0.84)
- EdgeMovingNet: Edge-preserving Point Cloud Reconstruction via Joint Geometry Features (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.
