Research map / Graph learning / Graph self-supervised learning
Graph Transformer GANs for Graph-Constrained House Generation
Hao Tang, Zhenyu Zhang, Humphrey Shi, Bo Li, Ling Shao, Nicu Sebe, Radu Timofte, Luc Van Gool
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 |
|---|---|---|---|
| use GTGAN instead of HouseGAN++* | house layout generation · LIFULL HOME's, all groups | realism: 0.19 -> 0.25 | Improved |
| use GTGAN instead of HouseGAN++* | house layout generation · LIFULL HOME's, 10-12 rooms | FID: 10.1 -> 7.5 | Improved |
| use GTGAN instead of HouseGAN++* | house layout generation · LIFULL HOME's, 10-12 rooms | compatibility: 3.65 -> 2.63 | Improved |
| replace HouseGAN generator with GTGAN generator (ablation) | B1 HouseGAN baseline · LIFULL HOME's, 10-12 rooms | FID: 11.6 -> 10.3 | Improved |
| replace HouseGAN discriminator with GTGAN discriminator (ablation) | B1 HouseGAN baseline · LIFULL HOME's, 10-12 rooms | FID: 11.6 -> 9.5 | Improved |
| use both GTGAN generator and discriminator (ablation) | B1 HouseGAN baseline · LIFULL HOME's, 10-12 rooms | FID: 11.6 -> 8.2 | Improved |
| add non-connected node attention (NNA) (ablation) | GTGAN · LIFULL HOME's, 10-12 rooms | FID: 9.4 -> 8.2 | Improved |
| add connected node attention (CNA) (ablation) | GTGAN · LIFULL HOME's, 10-12 rooms | FID: 9.7 -> 8.2 | Improved |
| add graph modeling block (GMB) (ablation) | GTGAN · LIFULL HOME's, 10-12 rooms | FID: 9.5 -> 8.2 | Improved |
| use Conv-MPN layers instead of Transformer layers (ablation) | GTGAN · LIFULL HOME's, 10-12 rooms | FID: 8.9 -> 8.2 | Improved |
| use the full Conv-MPN update instead of the Eq. [ref] variation (ablation) | GTGAN · LIFULL HOME's, 10-12 rooms | FID: 9.4 -> 8.2 | Improved |
| use the full Conv-MPN update instead of the other Eq. [ref] variation (ablation) | GTGAN · LIFULL HOME's, 10-12 rooms | FID: 9.8 -> 8.2 | Improved |
Closest papers by meaning
- Graph Transformer GANs With Graph Masked Modeling for Architectural Layout Generation (TPAMI 2024 · similarity 0.97)
- House-GAN: Relational Generative Adversarial Networks for Graph-constrained House Layout Generation (ECCV 2020 · similarity 0.92)
- House-GAN++: Generative Adversarial Layout Refinement Network towards Intelligent Computational Agent for Professional Architects (CVPR 2021 · similarity 0.88)
- Roof-GAN: Learning To Generate Roof Geometry and Relations for Residential Houses (CVPR 2021 · similarity 0.88)
- GraphGAN: Graph Representation Learning With Generative Adversarial Nets (AAAI 2018 · similarity 0.86)
- Generative Graph Pattern Machine (NeurIPS 2025 · similarity 0.83)
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.
