Çeşitli mimari temsil tekniklerinin üretilmesinde metinden görüntüye araçların yararlılığının değerlendirilmesi

Over time, the field of architecture has seen and passed by various technologies, from the simple usage of computers to the generative AI era that we live nowadays. Text-to-image AI is a current state-of-the-art technology, that is revolutionizing today's life, by being used in many domains and for various purposes. Architecture is one of the realms that has been adopting this technology recently. This thesis aims to explore the use of text-to-image tools in the field, by investigating their potential for shaping one of the basics of architectural design, which is architectural representational techniques, through a comparative study based on rating, by generating various architectural representational techniques using five different text-to-image tools available online, and presented to architects raters to rate the different generated images by these tools for the different representational techniques. The results of this study show the different aspects of text-based image-generation tools and how they can handle the generation of the representational techniques differently and propose the use of a multi-tool approach for better results while considering the variety of the generated images, which provides valuable insight into the use of these technologies in a more efficient way and can also shape the basics for future studies for deeper discoveries.