Generative AI allows the creation of any image based simply on text input. For this issue of Touchpoint, students from the Köln International School of Design created alternative covers using several models and methods. The text-to-image model is a machine learning model. It uses two neural networks: one creates the image, and the other evaluates, based on real examples of the target image, how close it is to the real image. After estimating the accuracy of the image, it sends that information back to the system and shows it to the user. On the these pages you can see examples of Touchpoint covers created using three distinct methods:
Become an SDN member today to start reading this article.
Discover our membership options
Become a Member