Workshop on New Trends in Optimization for Imaging

The workshop aims at bringing together International and Chinese researchers, scientists and graduate students to exchange and stimulate ideas in imaging sciences, with a special focus on new approaches based on optimization methods. Variational models are getting popular for image processing and other imaging problems include inverse problems, image construction and computer vision. However, it is not easy to get fast and robust “enough” algorithms to solve the variational models due to complexity involved with: 1) nonlinearity, 2) non-smoothness of functional and solutions, 3) non-convex for the minimization, 4) higher order derivatives within the functional, and 5) large data size. Recent researches have revealed that a number of numerical techniques can produce real time simulations and produce robust repeatable analysis tools. These new techniques involve techniques from a number of research areas including optimization theory, geometry, machine learning, combinatorial mathematics, PDE (partial differential) theory and numeric which is making this new area a real inter-disciplinary research field. New models and new numerical techniques originated from these interdisciplinary researches are coming into this field with promising industrial applications in recent years. This leads to the necessity to have a workshop like this one to bring researchers from different parts of the word and from different disciplines to get together to discuss new emerging results and to explore new possibilities. For this workshop, we will put emphasis to include techniques related to: 1) singularities, 2) Optimization techniques, such as the primal-dual method and the augmented Lagrangian methods, 3) Convex relaxation methods and global minimization, 4) combinatorial and graph related techniques 4) differential geometry for imaging. These are cutting-edge topics with crucial impact in various areas of imaging sciences (including, e.g. inverse problems, image processing and computer vision).

For the more workshop photos downloading, please refer to and


Raymond H. ChanChinese University of Hong Kong
Tony ChanHong Kong University of Science and Technology
Mila NikolovaCNRS, ENS Cachan, France
Xue-Cheng TaiUniversity of Bergen, Norway