This talk addresses opportunities as well as challenges connected to bilevel approaches in image restoration or blind deconvolution. While a major advantage is a monolithic optimization framework favoring leader-follower principles, some of the main challenges are related to non-smoothness, non- convexity, lack of constraint qualification as well as the design of efficient solution algorithms. Particular applications highlighted in this talk comprise distributed regularization parameter choice rules as well as topics in multiframe blind deconvolution.

This presentation is part of Minisymposium “MS5 - Learning and adaptive approaches in image processing (2 parts)”
organized by: Kostas Papafitsoros (Weierstrass Institute Berlin) , Michael Hintermüller (Humboldt University and Weierstrass Institute Berlin) .

Authors:

Michael Hintermüller (Humboldt University and Weierstrass Institute Berlin)

Keywords:

image deblurring, image reconstruction, inverse problems, nonlinear optimization