Research map / 3D vision / 3D generation
LayoutNet: Reconstructing the 3D Room Layout From a Single RGB Image
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 vanishing-point image alignment (ablation) | LayoutNet · PanoContext | 3D IoU (%): 69.91 -> 74.48 | Helped |
| add vanishing-point image alignment (ablation) | LayoutNet · Stanford 2D-3D annotation dataset | 3D IoU (%): 68.56 -> 75.39 | Helped |
| add joint boundary prediction to corner prediction (ablation) | LayoutNet · PanoContext | 3D IoU (%): 73.16 -> 73.26 | Helped |
| add the 3D layout loss (ablation) | LayoutNet corner+boundary · PanoContext | 3D IoU (%): 73.26 -> 74.48 | Helped |
| add the 3D layout loss (ablation) | LayoutNet corner+boundary · Stanford 2D-3D annotation dataset | 3D IoU (%): 75.26 -> 75.39 | Helped |
| add the cuboid constraint (ablation) | LayoutNet · PanoContext | 3D IoU (%): 72.56 -> 74.48 | Helped |
| add layout optimization (ablation) | LayoutNet · PanoContext | 3D IoU (%): 73.25 -> 74.48 | Helped |
| use cross-entropy rather than L2 loss (ablation) | LayoutNet · PanoContext | 3D IoU (%): 73.55 -> 74.48 | Helped |
| use cross-entropy rather than L2 loss (ablation) | LayoutNet · Stanford 2D-3D annotation dataset | 3D IoU (%): 75.39 -> 76.33 | Hurt |
| train on combined PanoContext and Stanford 2D-3D data (ablation) | LayoutNet full · PanoContext | 3D IoU (%): 74.48 -> 75.12 | Helped |
| train on combined Stanford 2D-3D and PanoContext data (ablation) | LayoutNet full · Stanford 2D-3D annotation dataset | 3D IoU (%): 75.39 -> 77.51 | Helped |
| use LayoutNet rather than PanoContext | Panorama cuboid layout estimation · PanoContext | 3D IoU (%): 67.23 -> 74.48 | Helped |
Closest papers by meaning
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction (NeurIPS 2026 · similarity 0.91)
- No More Ambiguity in 360deg Room Layout via Bi-Layout Estimation (CVPR 2024 · similarity 0.88)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.88)
- Unposed Sparse Views Room Layout Reconstruction in the Age of Pretrain Model (ICLR 2025 · similarity 0.87)
- Unified Panoramic Geometry Estimation via Multi-View Foundation Models (NeurIPS 2026 · similarity 0.86)
- CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction (NeurIPS 2025 · similarity 0.86)
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.