概念圖像的產生往往由具有豐富經驗的建築師，將腦中的思緒透過設計工具，具體成像在紙上或其他的設計介面來和其他設計團隊內的設計者或建築師進行設計深化。因此，本研究提出運用深度學習模型的生成技術建立一個影像溝通模型(Image Communicational Model, ICM)作為轉譯設計案業主之意圖的替代方式，並以認知易用性與認知有用性評估其溝通模型的可行性。業主有機會透過影像溝通模型傳達他們對設計的看法，進而激發設計者的設計靈感，發掘更多設計可能性，也讓設計者在和業主溝通的過程中有更多的彈性發展設計。研究中加入實驗對象參與設計並以訪談和線
上問卷作為收集資料的途徑。本研究認為影像溝通模型(ICM)可以幫助使用者探索設計的可能，提高對建築設計的關注與認知，並促進在設計初期與設計者的溝通。此外，研究結果表明，在某些狀況下，ICM 流程讓實驗對象對於設計成果有更高的滿意度，同時為設計者提供在設計面更大的自由度。影像不僅僅是做為溝通工具，同時更進一步解放了設計者和業主的在設計發想上的限制，電腦的生成能力在這當中扮演的重要的媒介。以電腦運算邏輯生成設計依然在非常前期而模糊的階段，但其未來的發展無疑將帶來建築設計專業的革命性轉變。

In a typical architectural design workflow, a design director collects informationgathered from a client and derives a formal concept based on his or her understanding of the project. Often, such formal concepts are expressed through representational medium such as drawings or models to other designers and architects in the office to develop into architectural plans. The purpose of this research is to develop an ImageCommunicational Model (ICM) using deep learning tools, which can provide an alternative method of translating client intent, and evaluate its ease of use and usefulness in such architectural design studios. This research project invited potential
clients and designers to participate in the design process, and conducted interviews and surveys as a mean to collect and organize data. Through this experiment, the ICM process was found to help users explore design possibilities, improve the attention and cognition of architectural designers, and facilitate communication with the designers during the preliminary design stage. Additionally, findings suggest that in some cases the ICM process was able to cultivate higher result satisfaction among participants, while providing greater design freedom to designers. Though the implementation of
computational design logic is still nascent in the field of architecture, its future development will undoubtedly yield a revolutionary shift in architectural design profession.

中文文獻
邱茂林 設計運算向度, 臺北：田園城市, 2003
Kelly K. 必然:掌握型塑未來30 年的12 科技大驅力, 嚴麗娟譯，臺北：貓頭鷹, 2017
Berge John 觀看的方式, 吳莉君譯，台北：麥田, 2010
Mitchell W. J. 位元城市, 陳瑞清譯，台北：天下文化, 1998
Lindsey B. 數位蓋瑞-探索材料極限的數位化建構, 宋偉祥譯，台北：旭營文化, 2003
Russell S. &Norvig P. 人工智慧現代方法, 高超群譯, 臺北:全華圖書, 2011
C. Norberg- Schulz 建築意向, 曾旭正譯,台北：胡式圖書, 民79
彭聃齡、張必隱 認知心理學, 台北：台灣東華, 2000
Mitchell W. J. 建築的設計思考, 劉育東譯,台北:建築情報出版：胡式圖書., 1995
殷晓蓉 網路傳播文化, 北京：清華大學, 2005
Wright F. 建築之夢, 于潼譯, 台北：博雅書屋, 2012
Dewey J. 藝術即經驗, 高建平譯, 北京：商務印書館, 2005
陳超萃 設計認知-設計中的認知科學, 北京：中國建築工業出版社, 2008

外文文獻
Aleksander A."Evolution of Computer Aided Design: Three Generations
of CAD." In Architectural Computing from Turing to 2000: 17th eCAADe
Conference ,1999 Proceedings, 94-100. eCAADe: Conferences. University of Liverpool, UK: University of Liverpool.
Alec R., Luke M., Soumith C. Unsupervised representation learning with deep
convolutionl generative adversarial networks . arXiv:1511.06434v2[cs.LG] 7 Jan., 2016
Creswellx A., White T., Dumoulinz V., Arulkumaranx K., Senguptayx B., Bharath A.. Generative Adversarial Networks: An Overview. arXiv:1710.07035v1 [cs.CV] 19 Oct., 2017
Carpo M., The second digital turn. Cambridge:The MIT Press, 2017
Carpo M.,. The digital turn in Architecture. New York: John Wiley & Sons Inc, 2012
Davis, Fred D., Bagozzi, Richard P. and Warshaw, Paul R. User acceptance of computer technology: a comparison of two theoretical models. Management Science, 35(8), 1989,982-1003.
Davis, F.D. “Perceived usefulness, perceived ease of use, and user acceptance
ofinformation technology”, MIS Quarterly, Vol. 13, No. 3, 1989, pp. 319-340.
Gero J.S., Design Prototype : A knowledge representation schema for design. AI Magazine, Vol.11, Number 4, 1990
Ian Goodfellow. NIPS 2016 Tutorial: Generative Adversarial Networks. arXiv:1701.00160v4[cs.LG] 3 Apr., 2017
Krizhevsky A., Sutskever I., Hinton G.E. ImageNet Classification with Deep Convolutional Neural Networks NIPS'12 Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1 ,2012, Pages 1097-1105,
Lynn G., Animate Form, New York:Princeton Architectural Press.,1999
Rappolt M., Greg Lynn form. New York:Rizzoli International Publications, 2008
Oxman.R. The thinking eye: visual re-cognitionin design emergence, Design Studies Vol 23 No. 2 March 2002
Minsky, M The society of the mind ,Simon & Schuster, New York ,1986,
Margolius I.., Painting as Architectural storyboards: Zaha Hadid in Conversation with Ivan Margolius, Architectural Designer, vol. 73, no. 3, England: Wiley Academy2003.
Negroponte, Nicholas. The Architecture Machine: Toward a More Human Environment. Cambridge:The MIT Press, 1973
Radford A., Metz L., Chintala S., Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. arXiv:1511.06434 [cs.LG],2016
Radford A., Metz L., Chintala S. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, arXiv:1511.06434 [cs.LG], 2016
Ståhla M., Kaihovirtab H. Exploring visual communication and competencies through interaction with images in social media. Learning, Culture and Social Interaction,Volume 21, June 2019, Pages 250-266
Turing A . M., Computing Machinery and Intelligence. Mind, New Series, Vol. 59, No. 236 Oxford University Press on behalf of the Mind Association, Oct., 1950, pp. 433-460.
Yehuda E. Kalay. Architecture's New Media: Principles, Theories, and Methods of Computer-Aided Design. Cambridge:The MIT Press, 2004
Zhang H. etc StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks arXiv:1612.03242 [cs.CV], 2017

重要會議:
NIPS神經信息處理系統大會 Conference and Workshop on Neural Information Processing Systems

網路資料來源
李宏毅，ML Lecture 17: Unsupervised Learning - Deep Generative Model (Part I), 2016
https://www.youtube.com/watch?v=YNUek8ioAJk
carpedm20，Github上的教學頁面(DCGAN-tensorflow),2016
https://github.com/carpedm20/DCGAN-tensorflow
Achimmenges個人頁面
http://www.achimmenges.net/
Patrik Schumacher 著作線上收錄
https://www.patrikschumacher.com/index.htm
GAN ZOO
https://github.com/hindupuravinash/the-gan-zoo/blob/master/gans.tsv
蔡炎龍，函數、神經網路與深度學習，科學月刊, 2019
http://scimonth.blogspot.com/2018/03/blog-post_8.html

應用軟體 資源平台
Anaconda https://www.anaconda.com/
GitHub https://github.com/