Computational Imaging & Learning

The school introduces the mathematical and computational foundations of computational imaging, emphasizing the modeling, formulation, and solution of inverse problems in modern imaging systems. It begins with physical models of image acquisition in contexts such as microscopy and tomography, which are discretized into ill-posed inverse problems. To address these, the course explores regularization and Bayesian methods like maximum a posteriori (MAP) estimation. Optimization techniques for image reconstruction, including gradient-based and proximal algorithms, are then presented. The course concludes with data-driven and learning-based methods that integrate with model-based approaches to enhance performance and interpretability. Applications include super-resolution microscopy, tomographic reconstruction, and medical imaging.

Courses span five days, with lectures held in the mornings and afternoons, except on Wednesday, which includes presentations given by the attendees and some invited seminar. Attendance is required and will be registered before each session. Students will have the chance to take a final exam (modalities TBC). Students will receive an attendance certificate indicating the number of hours they attended, provided that it is no less than 80% of the total course hours. The school will take place exclusively in person, it will not be streamed online nor recorded. Please note that there is no scholarship available to cover travel and accommodation costs.

The course covers theoretical and numerical aspects on the topic, it is targeted to students with solid background in applied mathematics and computer science, particularly in advanced calculus, linear algebra, numerical analysis and scientific calculus. Basic knowledge of Python is required. Lab classes will be given based on the DeepInverse library.