Publication 19-CNA-001

Adaptive Image Processing: First Order PDE Constraint Regularizers And A Bilevel Training Scheme

Elisa Davoli
Institute of Analysis and Scientific Computing
TU Wien
Wiedner Hauptstrasse 8-10
1040 Vienna, Austria
elisa.davoli@tuwien.ac.at

Irene Fonseca
Department of Mathematical Sciences
Carnegie Mellon University
Pittsburgh, PA 15213
fonseca@andrew.cmu.edu

Pan Liu
Department of Radiology
First Medical Center of Chinese PLA General Hospital
Beijing 100853, China
dragonrider.liupan@gmail.com

Abstract: A bilevel training scheme is used to introduce a novel class of regularizers, providing a unified approach to standard regularizers TGV² and NsTGV². Optimal parameters and regularizers are identified, and the existence of a solution for any given set of training imaging data is proved by $\Gamma$-convergence. Explicit examples and numerical results are given.

Get the paper in its entirety as  19-CNA-001.pdf

---

«   Back to CNA Publications