Exploring the photorealistic capabilities of AI image generators: Stable Diffusion, DALL-E mini and dream by WOMBO

Aleksandra Bernašek Petrinec, Ivan Papiga, Katja Milković, Silvio Plehati

Abstract

In today’s digital age, artificial intelligence plays a key role in transforming the way we create and perceive art and photography. Now it is possible to generate imagined visual content with input of just a few words of text. Rapid technological advancement has led to significant progress in the field of 
artificial intelligence, which is increasingly being used in various aspects of human life. For recent years, AI tools have become progressively popular in academic circles, particularly for testing various algorithms in the creation of scientific and professional experiments. This paper explores the artificial intelligence generators Stable Diffusion, DALL-E mini, and Dream by WOMBO to study the hypothesis whether high-quality photorealistic images can be created with simple input, without additional iterations.