Parametric Design Method of Buildings Based on Convolutional and Deep Neural Networks
DOI:
https://doi.org/10.61173/98vn3p38Keywords:
Architectural Design, Convolutional Neural Network, Deep Neural Network, Fusion Application, Research ProgressAbstract
The application of AI technology in architectural software has brought various innovations and changes to the industry. By introducing convolutional neural networks and deep neural networks, architectural design software is expected to automatically identify and generate complex architectural images, optimize the design process, and provide intelligent design assistance. Through literature research, this paper aims to discuss the specific application of CNN and DNN in parametric design and the innovation brought by them, and analyzes the application of these technologies in architectural image recognition, generation, design optimization and other aspects. This paper aims to provide a comprehensive reference for researchers in related fields and promote the innovative application of AI technology in architectural design. The application of AI technology in architectural software has brought numerous innovations and changes to the industry. By introducing convolutional neural networks and deep neural networks, architectural design software is expected to automatically identify and generate complex architectural images, optimize the design process, and provide intelligent design assistance. Through literature research, this paper aims to discuss the specific application of CNN and DNN in parametric design and the innovation brought by them; it also analyzes the application of these technologies in architectural image recognition, generation, design optimization, and other aspects. This paper aims to provide a comprehensive reference for researchers in related fields and promote the innovative application of AI technology in architectural design.
References
[1] Zhou Feiyan, Jin Linpeng, Dong Jun. Review of In terms of user customization, deep neural networks proconvolutional neural networks [J]. Chinese Journal of Compute vide strong support for designers. By learning from users’ rs,2017,40(06):1229-1251. preferences and needs, DNN can generate deeply per-
[2] Zheng Yuanpan, Li Guangyang, Li Ye. Survey of Application sonalized design solutions that meet the particular needs of Deep Learning in Image Recognition [J]. Computer of different users. In interior design, DNN can generate Dean&Francis Engineering and Applications, 2019, 55 (12): 20-36. Management, 2022, 148(11): 03122009.
[3] Lomio F, Farinha R, Laasonen M, et al. Classification [6] S Guo,Ren W Q, Qu Y B, Dong C, et al. A survey on of building information model (BIM) structures with deep collaborative DNN inference for edge intelligence[J]. Machine learning[C]//2018 7th European Workshop on Visual Information Intelligence Research, 2023, 20(3): 370-395. Processing (EUVIP). IEEE, 2018: 1-6. [7] Liu Gengzhe, Liu Xiaowen, Gou Zhaoyuan.Future
[4] Alawadhi M, Yan W. BIM hyperreality: Data synthesis using Architectural Scene Design based on Virtual Reality(VR)/ BIM and hyperrealistic rendering for deep learning[J]. arXiv Augmented Reality (AR) Technology:A Case from Simon Kim’s preprint arXiv:2105.04103, 2021. Studio of University of Pennsylvania [J]. Architecture and
[5] [5]Asghari V, Wang Y, Biglari A J, et al. Reinforcement culture, 2022 (8) : 34-36. DOI: 10.19875 / j.carol carroll nki. [17] learning in construction engineering and management: jzywh. 2022.08.011. A review[J]. Journal of Construction Engineering and
Downloads
Published
Issue
Section
License
Copyright (c) 2024 by the authors.
This work is licensed under a Creative Commons Attribution 4.0 International License.
