Regularization in Image Reconstruction: From Model to Data Driven Methods

This course addresses key mathematical questions and common challenges in image reconstruction problems. It provides the theoretical foundation of inverse problems and explains how different imaging modalities and measurement errors can affect the quality of reconstructions.

Regularization strategies are presented that help to overcome these difficulties, both from a theoretical and a practical perspective. The knowledge gained serves as a basis for continued self-guided learning during and after the course.

Building on classical regularization approaches, the course also offers an introduction to deep learning methods for inverse problems. In addition, the Bayesian approach is discussed, including the topic of uncertainty quantification.

The workshop consists of a lecture and a practical session with exercises. Throughout the course, the instructors are available for questions, feedback, and individual support.