Chen Junming, Shao Zichun, Zheng Xiaodong, Zhang Kai, Yin Jun
Faculty of Humanities and Arts, Macau University of Science and Technology, Taipa, 999078, Macau.
School of Design, Jiangnan University, Wuxi, 214000, China.
Sci Rep. 2024 Feb 12;14(1):3496. doi: 10.1038/s41598-024-53318-3.
The interior design suffers from inefficiency and a lack of aesthetic appeal. With the development of artificial intelligence diffusion models, using text descriptions to generate aesthetically pleasing designs has emerged as a new approach to address these issues. In this study, we propose a novel method based on the aesthetic diffusion model, which can quickly generate visually appealing interior design based on input text descriptions while allowing for the specification of decorative styles and spatial functions. The method proposed in this study creates creative designs and drawings by computer instead of from designers, thus improving the design efficiency and aesthetic appeal. We demonstrate the potential of this approach in the field of interior design through our research. The results indicate that: (1) The method efficiently provides designers with aesthetically pleasing interior design solutions; (2) By modifying the text descriptions, the method allows for the rapid regeneration of design solutions; (3) Designers can apply this highly flexible method to other design fields through fine-tuning. (4) The method optimizes the workflow of interior design.
室内设计存在效率低下和缺乏审美吸引力的问题。随着人工智能扩散模型的发展，利用文本描述生成美观的设计已成为解决这些问题的一种新方法。在本研究中，我们提出了一种基于审美扩散模型的新颖方法，该方法可以根据输入的文本描述快速生成视觉上吸引人的室内设计，同时允许指定装饰风格和空间功能。本研究提出的方法通过计算机而非设计师来创建创意设计和图纸，从而提高了设计效率和审美吸引力。我们通过研究证明了这种方法在室内设计领域的潜力。结果表明：（1）该方法有效地为设计师提供了美观的室内设计解决方案；（2）通过修改文本描述，该方法允许快速重新生成设计解决方案；（3）设计师可以通过微调将这种高度灵活的方法应用于其他设计领域。（4）该方法优化了室内设计的工作流程。
