Research map / 3D vision / 3D generation
HouseCrafter: Lifting Floorplans to 3D Scenes with 2D Diffusion Models
Yiwen Chen, Hieu T. Nguyen, Vikram Voleti, Varun Jampani, Huaizu Jiang
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 depth output (generate RGB-D jointly) (ablation) | RGB-only diffusion model · 3D-FRONT RGB-D ablation (300 houses) | FID: 49.35 -> 33.39 | Improved |
| add depth output (generate RGB-D jointly) (ablation) | RGB-only diffusion model · 3D-FRONT RGB-D ablation (300 houses) | IS: 5.00 -> 5.23 | Improved |
| add input reference depth (ablation) | RGB-D model without floorplan · 3D-FRONT RGB-D ablation (300 houses) R-N | AbsRel: 23.56 -> 22.28 | Improved |
| add input reference depth (ablation) | RGB-D model with floorplan · 3D-FRONT RGB-D ablation (300 houses) R-N | AbsRel: 7.65 -> 7.73 | Hurt |
| add input reference depth (ablation) | RGB-D model without floorplan · 3D-FRONT RGB-D ablation (300 houses) | FID: 33.39 -> 35.77 | Hurt |
| add floorplan conditioning (ablation) | RGB-D model · 3D-FRONT RGB-D ablation (300 houses) | FID: 33.39 -> 15.64 | Improved |
| add floorplan conditioning (ablation) | RGB-D model · 3D-FRONT RGB-D ablation (300 houses) R-N | AbsRel: 23.56 -> 7.65 | Improved |
| add floorplan conditioning (ablation) | RGB-D model · 3D-FRONT RGB-D ablation (300 houses) | IS: 5.23 -> 4.70 | Hurt |
| add floorplan conditioning (ablation) | RGB-D model with input depth · 3D-FRONT RGB-D ablation (300 houses) | FID: 35.77 -> 16.70 | Improved |
| replace CaPE with DeCaPE (ablation) | RGB-D model with floorplan · 3D-FRONT RGB-D ablation (300 houses) R-N | AbsRel: 7.73 -> 6.79 | Improved |
| replace CaPE with DeCaPE (ablation) | RGB-D model with floorplan · 3D-FRONT RGB-D ablation (300 houses) | mAP@25: 49.74 -> 52.26 | Improved |
| replace CaPE with DeCaPE (ablation) | RGB-D model with floorplan · 3D-FRONT RGB-D ablation (300 houses) | FID: 16.70 -> 17.75 | Hurt |
Closest papers by meaning
- SceneCraft: Layout-Guided 3D Scene Generation (NeurIPS 2024 · similarity 0.86)
- DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (CVPR 2024 · similarity 0.86)
- RoomPainter: View-Integrated Diffusion for Consistent Indoor Scene Texturing (CVPR 2025 · similarity 0.86)
- A 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale (NeurIPS 2026 · similarity 0.85)
- Channel-wise Noise Scheduled Diffusion for Inverse Rendering in Indoor Scenes (CVPR 2025 · similarity 0.85)
- Diffusion Renderer: Neural Inverse and Forward Rendering with Video Diffusion Models (CVPR 2025 · 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.
