Research map / Point clouds / Shape and surface completion
MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans
Sinisa Stekovic, Mahdi Rad, Friedrich Fraundorfer, Vincent Lepetit
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 the refinement step (ablation) | MonteFloor · Structured3D | MA precision / recall: 0.82 -> 0.90 precision, 0.75 -> 0.82 recall | Improved |
| add the refinement step (ablation) | MonteFloor · Structured3D | Angle precision: 0.65 -> 0.86 | Improved |
| add the refinement step (ablation) | MonteFloor · Structured3D | computation time (s): 12 ± 8 -> 71 ± 40 | Hurt |
| add angle regularization L_ang (ablation) | MonteFloor · Structured3D | Angle precision: 0.73 -> 0.86 | Improved |
| add angle regularization L_ang (ablation) | MonteFloor · Structured3D | Angle recall: 0.68 -> 0.75 | Improved |
| add global contact regularization L_glob (ablation) | MonteFloor · Structured3D | Room precision: 0.85 -> 0.96 | Improved |
| add global contact regularization L_glob (ablation) | MonteFloor · Structured3D | Corner precision: 0.78 -> 0.89 | Improved |
| add drift regularization L_0 (ablation) | MonteFloor · Structured3D | MA precision: 0.88 -> 0.90 | Improved |
| add drift regularization L_0 (ablation) | MonteFloor · Structured3D | Room precision: 0.92 -> 0.96 | Improved |
| add the learned metric network f (ablation) | MonteFloor · Structured3D | MA recall: 0.17 -> 0.82 | Improved |
| replace Floor-SP with MonteFloor (MCTS selection and refinement) | floor plan reconstruction · Structured3D | MA precision / recall: 0.83 -> 0.90 precision, 0.78 -> 0.82 recall | Improved |
| replace Floor-SP with MonteFloor, metric network trained only on Structured3D | floor plan reconstruction · Floor-SP test set | MA precision / recall: 0.74 -> 0.78 precision, 0.64 -> 0.67 recall | Improved |
Later papers that compared with this method
1 later accepted paper reports MonteFloor as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SLIBO-Net: Floorplan Reconstruction via Slicing Box Representation with Local Geometry Regularization (NeurIPS 2023 · better on 4 of 4 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
- MCTS With Refinement for Proposals Selection Games in Scene Understanding (TPAMI 2022 · similarity 0.95)
- Monte Carlo Scene Search for 3D Scene Understanding (CVPR 2021 · similarity 0.90)
- Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries (CVPR 2023 · similarity 0.88)
- CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction (NeurIPS 2025 · similarity 0.87)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.86)
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction (NeurIPS 2026 · similarity 0.85)
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.
