The mathematics behind generative AI models in imaging

Lecturer: Martin Holler, University of Graz

Generative AI models have revolutionized computational imaging. They enable unprecedented capabilities in consumer photography and visual content creation, and have also become the state of the art in scientific applications such as biomedical imaging.

In this talk, Martin Holler will explain the mathematics behind generative AI models in imaging. You will discuss their probabilistic foundations and main working principles, potential pitfalls, how they can be tuned to specific imaging problems, and illustrate these ideas with real-world examples from Magnetic Resonance Imaging.