Abstract
Architecture does not exist in a vacuum. Its culture, concepts and aesthetics are constantly influenced by other visual and artistic disciplines, from film, photography, painting and sculpture to fashion, graphic and industrial design. In the field of architecture, design generation work requires architects to collect a lot of data and understand many details before they can design effectively. In the era of big data, design work is becoming more and more complex, and architects are increasingly required to effectively learn and manage the data in front of them, and the process of learning data can actually be done through machine learning. As a generative tool that can effectively learn a large amount of data, the generative deep learning model has played a good role in some architectural design generation work. The main goal of this research is to find inspiration for new aesthetic paradigms outside the architectural discipline. The experimental architectural landmark selection of this paper selects the buildings in the center of Qingdao, China, as our experimental site. Through deep learning and neural network, aiming at the content images of architectural elements of interest, while maintaining the architectural characteristics of the content images, they are stylized with artistic images to provide effective assistance for the facade design of modern urban high-rise buildings.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call