Martin Holler

Martin Holler is a Professor at the Institute of Mathematics and Scientific Computing at the University of Graz, Austria, where he leads the research group Applied Mathematics and Machine Learning. His work bridges theoretical mathematics with practical applications in imaging and machine learning.

Research Focus: His primary research areas include the mathematics of data science, variational methods in imaging, dynamic and multi-modality inverse problems, and biomedical imaging. He has made significant contributions to model-based regularization techniques, particularly with Total Generalized Variation (TGV) approaches for image and video reconstruction.

Publication Trends: Over the past decade, Holler's research has evolved from traditional variational methods for image reconstruction toward increasingly sophisticated machine learning approaches. His recent work (2021-2023) focuses on integrating deep learning with variational methods, particularly for motion separation in medical imaging and learning-informed parameter identification in partial differential equations. His publications demonstrate a consistent thread of applying rigorous mathematical frameworks to solve practical problems in medical imaging and computer vision.

• Mathematics of data science and machine learning
• Generative models in machine learning
• Variational methods in imaging
• Dynamic and multi-modality inverse problems
• Model-based regularization
• Biomedical imaging
• Image and video decompression

Technical Leadership: Holler has developed several open-source software packages implementing advanced reconstruction algorithms, particularly for multi-modal imaging problems. His GitHub repositories show active maintenance and development of these tools, which have been cited in the medical imaging community.