Research map / 3D vision / 3D generation
ArchiSet: Benchmarking Editable and Consistent Single-View 3D Reconstruction of Buildings with Specific Window-to-Wall Ratios
Jun Yin, Pengyu Zeng, Licheng Shen, Miao Zhang, Jing Zhong, Yuxing Han, Shuai Lu
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 window-region consistency regularization loss to multi-view generation (ablation) | BuildMesh · ArchiSet multi-view generation | FID: 2.846 -> 1.833 | Improved |
| add window-region consistency regularization loss to multi-view generation (ablation) | BuildMesh · ArchiSet multi-view generation | PSNR: 14.875 -> 16.374 | Improved |
| add window-region consistency regularization loss to multi-view generation (ablation) | BuildMesh · ArchiSet multi-view generation | LPIPS: 0.326 -> 0.238 | Improved |
| use BuildMesh instead of PC2 for single-view building generation | ArchiSet WS · ArchiSet WS (shading diagrams) | F-score: 0.6740 -> 0.7396 | Improved |
| use BuildMesh instead of One-2-3-45 on RealCity3D | RealCity3D · RealCity3D | Chamfer Distance: 0.1072 -> 0.1087 | Hurt |
| use BuildMesh instead of One-2-3-45 on BuildingNet | BuildingNet · BuildingNet | F-score: 0.4694 -> 0.4648 | Hurt |
| condition generation on window-to-wall ratio | BuildMesh · ArchiSet WS/WR/WL | WWR relative error: 3.24%, 8.89%, 9.32% on WS, WR, WL | Improved |
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.
- One-2-3-45 — One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization (NeurIPS 2023) · better on 4 of 6 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
- Visual-to-Executable Procedural Reconstruction of Buildings. (NeurIPS 2026 · similarity 0.88)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.88)
- BuildingNet: Learning To Label 3D Buildings (ICCV 2021 · similarity 0.87)
- ArcPro: Architectural Programs for Structured 3D Abstraction of Sparse Points (CVPR 2025 · similarity 0.85)
- OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization (NeurIPS 2025 · similarity 0.85)
- GSDiff: Synthesizing Vector Floorplans via Geometry-enhanced Structural Graph Generation (AAAI 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.
