In the first half of this thesis, I describe the process of image generation using artificial
intelligence and explain the basics of how these programs work. I also discuss current
questions about the authorship of images generated with the help of artificial intelligence,
questions about the continued relevance of illustrators (will artificial intelligence replace
our work?), and give my thoughts on how artificial intelligence can affect the work of
illustrators today and in the future.
In the practical part of the master's thesis, I investigate the competencies of artificial
intelligence programs in illustration with a comparative analysis. Using the programs
Stable Diffusion and Midjourney, I investigate how successfully these two programs
generate illustrations with prompts and reference image material. I then evaluate the
results on the basis of relevant parameters (consistency of artistic style and portrayal
of characters, production time of illustrations, visual narration, etc.), compare them with
my personal illustrations and therefore give my conclusion of the research.
|