返回
Human-guided urban form generation using multimodal diffusion models
DOI:10.1016/j.buildenv.2025.113892.png)
摘要
En 中文
• 采用多模态扩散模型开发了一种城市形态生成框架。
• 三阶段流程:道路和土地利用、建筑布局以及详细渲染。
• ControlNet 集成图像约束和文本提示以指导设计生成。
• 在合规性、保真度和多样性方面优于基线模型和端到端模型。
• 使人机迭代协作成为可能,用于适应性、情境敏感的城市设计。
Keyword:
Urban form generation
Multimodal diffusion models
Human-AI interaction
Building layout
Urban design
期刊
IF:
7.6
论文数:
1.3W
被引数:
6.6W
机构
引用论文
Cai, D. (2024). Enhancing capabilities of generative models through VAE-GAN integration: A review. Applied and Computational Engineering, 67, 239–246. https://www.ewadirect.com/proceedings/ace/article/view/14698. 10.54254/2755-2721/67/2024MA0070. TLDR: This review explores the integration of Variational Autoencoders and Generative Adversarial Networks (GANs), which are pivotal in the realm of generative models, and provides a comprehensive overview of VAEs and GANs technology developments, their integration strategies, and resultant performance improvements.蔡, D. (2024). 通过VAE-GAN集成提升生成模型能力:综述. 应用与计算工程, 67, 239–246. https://www.ewadirect.com/proceedings/ace/article/view/14698. 10.54254/2755-2721/67/2024MA0070. TLDR: 本文综述探讨了变分自编码器(Variational Autoencoders)和生成对抗网络(GANs)的集成,这两种技术在生成模型领域具有关键作用,并全面概述了VAEs和GANs的技术发展、集成策略及由此带来的性能提升。
Urban-GAN: An artificial intelligence-aided computation system for plural urban designUrban-gan: 用于多元城市设计的人工智能辅助计算系统
Alidoust, S., and Huang, W. (2022). Master Planned Communities for Healthy Living: A Systematic Literature Review. Urban Policy and Research, 40, 142–160. https://doi.org/10.1080/08111146.2022.2057945. 10.1080/08111146.2022.2057945.Alidoust, S., 和 Huang, W. (2022). 为健康生活而规划的社区:一项系统文献综述。城市规划与研究,40,142–160. https://doi.org/10.1080/08111146.2022.2057945. 10.1080/08111146.2022.2057945.
Deep Human-guided Conditional Variational Generative Modeling for Automated Urban Planning深度人引导条件变分生成建模用于自动化城市规划

