Research map / Point clouds / Shape and surface completion
FRI-Net: Floorplan Reconstruction via Room-wise Implicit Representation
Honghao Xu, Juzhan Xu, Zeyu Huang, Pengfei Xu, Hui Huang, Ruizhen Hu
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 two-phase training (axis-aligned lines first, then diagonal) instead of optimizing all lines together (ablation) | FRI-Net · Structured3D test | Corner precision: 76.1 -> 90.8 | Improved |
| add wall-height map to density map in pre-processing (ablation) | FRI-Net · Structured3D test | Corner precision: 88.1 -> 90.8 | Improved |
| increase lines per type from 256 to 1024 (ablation) | FRI-Net · Structured3D test | Corner precision: 90.8 -> 86.9 | Hurt |
| increase lines from 256 to 512 (ablation) | FRI-Net · Structured3D test | Corner recall: 84.9 -> 85.2 | No clear effect |
| increase convex primitives from 64 to 128 (ablation) | FRI-Net · Structured3D test | Corner precision: 90.8 -> 91.1 | No clear effect |
| reduce convex primitives from 64 to 32 (ablation) | FRI-Net · Structured3D test | Corner precision: 90.8 -> 89.5 | Hurt |
| room-wise implicit representation instead of corner sequence prediction | RoomFormer · Structured3D test | Angle F1: 81.2 -> 86.9 | Improved |
| room-wise implicit representation under train-on-Structured3D, test-on-SceneCAD | RoomFormer · Structured3D to SceneCAD (cross-data) | Corner F1: 60.3 -> 72.4 | Improved |
| FRI-Net versus RoomFormer on runtime | RoomFormer · Structured3D test | runtime (s): 0.01 -> 0.09 | Hurt |
| add PolyDiffuse post-optimization to FRI-Net | FRI-Net · Structured3D test | Corner F1: 87.8 -> 91.1 | Improved |
| FRI-Net versus RoomFormer on corner recall | RoomFormer · Structured3D test | Corner recall: 85.3 vs 84.9 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports FRI-Net as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction (NeurIPS 2025 · better on 0 of 1 shared result)
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.
- SLIBO-Net — SLIBO-Net: Floorplan Reconstruction via Slicing Box Representation with Local Geometry Regularization (NeurIPS 2023) · better on 3 of 3 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
- SLIBO-Net: Floorplan Reconstruction via Slicing Box Representation with Local Geometry Regularization (NeurIPS 2023 · similarity 0.92)
- Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries (CVPR 2023 · similarity 0.91)
- PolyRoom: Room-aware Transformer for Floorplan Reconstruction (ECCV 2024 · similarity 0.90)
- FloorNet: A Unified Framework for Floorplan Reconstruction from 3D Scans (ECCV 2018 · similarity 0.89)
- CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction (NeurIPS 2025 · similarity 0.88)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 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.
