An attempt to integrate AI-based techniques into first year design representation course

Abstract

This research aims to explore the potential application of AI-based techniques in the introduction of design representation during first-year education. Specifically, the study examines the integration of traditional representation techniques used in design education with AI algorithms through an assignment implemented in a course namely Visual Communication 1. This course is part of the Integrated Foundation Studio (IFC) model, which has been implemented at Istanbul Technical University, Faculty of Architecture for the past 8 years. The assignment was completed by 64 students from three different departments (architecture, interior architecture, industrial design) who were all freshmen with little prior knowledge and experience in design representation and digital tools. The assignment has three steps as: (i) creating a composition with basic solids and producing an isometric projection of the composition by hand drawing, (ii) generating a series of images by using an AI-based tool Midjourney by defining keywords, and (iii) combining the previous two productions using various techniques. As a result of this assignment, 50 out of 64 students produced and submitted their personalized and unique design representations. The outcomes of the assignment are qualitatively evaluated under two main categories as Combining Techniques and Atmospheric Elements. While the first main category Combining Techniques refers to types of used modification and visualization techniques, Atmospheric Elements deals with texture, color, and forms derived from AI-generated images. Initial findings indicate that text-to-image techniques of AI contribute to enhancing the interpretation and composition skills of freshmen while producing new visual representations by making use of source images and technical drawings.