Abstract
Traditional architectural design research is predominantly based on images and drawings of precedents from built projects or speculative theoretical models, in a process of critical analyses or points of departure for design iterations. Emerging AI visualization tools enable users to generate visual concepts and models that are derived from ever-evolving web data beyond pre-conceived and pre-categorized sources. This research project experiments AI tools in visualization that can negotiate the established boundaries of theoretical and practical design concepts in both literal and metaphorical dimensions. In “The Datafied Society: Studying Culture Through Data”, a ‘query design strategy’ has become more user-defined. A query normally begins with keywords which posit certain sets of notions, hints, or clues that enable researchers to find a cipher or code. A topic of architectural research can be further interpreted or defined as a sequence or a set of keywords. Through the inductive method of trials-and-errors, these AI visualization tools are capable to translate meanings and ideas into images, therefore, a wide range of images generated by AI can be selected to serve as visual interpretations and inspirations for architects and designers. As images and the meanings of images are also contextual, and the context and constraint of each research topic exercises can be systematically defined, which allow the outcomes to be compared and analyzed. Fundamentally, architectural design-based research is interpretive and creative. Humanistic AI visualization tools can generate both analogous and metaphoric images, which can be developed with great potentials in advancing architectural research methodology.
Presenters
Yong HuangAssistant Professor, Architecture, Bowling Green State University, Ohio, United States
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
KEYWORDS
AI VISUALIZATION, INTERPRETIVE RESEARCH, INDUCTIVE METHOD, LITERAL, METAPHORICAL
