Research map / 3D vision / 3D generation
GeoLayout: Geometry Driven Room Layout Estimation Based on Depth Maps of Planes
Weidong Zhang, Wei Zhang, Yinda Zhang
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 |
|---|---|---|---|
| predict pixel-level surface parameters and derive layout geometrically instead of PlaneNet's 2D segmentation | PlaneNet · Matterport3D-Layout test | pixel error e_pix: 6.89 -> 5.24 | Improved |
| use the geometric approach instead of PlaneNet | PlaneNet · Matterport3D-Layout test | 3D corner error: 14.00 -> 12.82 | Improved |
| use intrinsics-free surface parameters instead of standard plane parameters (ablation) | GeoLayout-Plane · Matterport3D-Layout test | depth rms: 0.448 -> 0.456 | No clear effect |
| use intrinsics-free surface parameters instead of standard plane parameters (ablation) | GeoLayout-Plane · Matterport3D-Layout test | pixel error e_pix: 5.84 -> 5.24 | Improved |
| use intrinsics-free surface parameters instead of standard plane parameters (ablation) | GeoLayout-Plane · Matterport3D-Layout test | 3D corner error: 12.05 -> 12.82 | Hurt |
| fine-tune on LSUN with 2D-only learning strategy (ablation) | GeoLayout trained on Matterport3D-Layout · LSUN validation | pixel error e_pix (%): 12.67 -> 6.10 | Improved |
| fine-tune on LSUN with 2D-only learning strategy (ablation) | GeoLayout trained on Matterport3D-Layout · LSUN validation | corner error e_cor (%): 8.12 -> 4.66 | Improved |
| incorporate 3D geometric reasoning (GeoLayout) versus best prior 2D method | Zhang et al. (2019) · LSUN test | pixel error e_pix (%): 6.58 -> 6.09 | Improved |
| incorporate 3D geometric reasoning (GeoLayout) versus best prior method | Hsiao et al. (2019) · Hedau | pixel error e_pix (%): 5.01 -> 7.16 | Hurt |
Later papers that compared with this method
1 later accepted paper reports GeoLayout as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- MCTS With Refinement for Proposals Selection Games in Scene Understanding (TPAMI 2022 · better on 5 of 5 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.
- PlaneNet — PlaneNet: Piece-Wise Planar Reconstruction From a Single RGB Image (CVPR 2018) · 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
- PolyLayout: Multi-room Manhattan Layout Estimation (ECCV 2026 · similarity 0.91)
- Spatial Geometric Reasoning for Room Layout Estimation via Deep Reinforcement Learning (ECCV 2020 · similarity 0.90)
- 3D Room Layout Estimation from a Cubemap of Panorama Image via Deep Manhattan Hough Transform (ECCV 2022 · similarity 0.90)
- RoomNet: End-To-End Room Layout Estimation (ICCV 2017 · similarity 0.90)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.89)
- No More Ambiguity in 360deg Room Layout via Bi-Layout Estimation (CVPR 2024 · similarity 0.88)
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.
