An Introduction to Total Variation for Image Analysis

These are the lecture notes of a course taught in Linz in Sept., 2009, at the school summer school on sparsity, organized by Massimo Fornasier and Ronny Romlau. They address various theoretical and practical topics related to Total Variation-based image reconstruction. They focu first on some theoretical results on functions which minimize the total variation, and in a second part, describe a few standard and less standard algorithms to minimize the total variation in a finite-differences setting, with a series of applications from simple denoising to stereo, or deconvolution issues, and even more exotic uses like the minimization of minimal partition problems.