Learning and adaptive approaches in image processingMS5

Learning and adaptive regularisation approaches have become popular in image processing. The complexity of modern imaging tasks, especially those in medical imaging have given rise to the need for more sophisticated, non-standard regularisations, where learning approaches are used to determine the selection of optimal parameters, forward models, data fitting terms or even the regularisation functionals. This minisymposium will bring together researchers with experience in the fields of parameter learning, non-standard adaptive and/or anisotropic approaches and their analysis - not necessarily in the context of regularisation - while particular emphasis will be given on medical imaging applications, e.g. Magnetic Resonance Imaging.

Tue 05 June at 13:30 in Room M (Palazzina B - Building B floor 0)
A physical-oriented reconstruction method for quantitative magnetic resonance imaging
Guozhi Dong ( Humboldt University Berlin )
A Variational Model for Brain Tumor Segmentation: Deep Learning Based Parameter Optimization
Adrián Martín ( Universitat Pompeu Fabra )
Analytical solutions of quadratic variational coupling models of arbitrary order
Aaron Wewior ( Saarland University )
Bilevel optimization and some "parameter learning" applications in image processing
Michael Hintermüller ( Humboldt University and Weierstrass Institute Berlin )
Tue 05 June at 16:00 in Room M (Palazzina B - Building B floor 0)
Deep regularization for medical image analysis
Erich Kobler ( Graz University of Technology )
Joint reconstruction and segmentation from undersampled MRI data
Veronica Corona ( University of Cambridge )
Learning a sampling pattern for MRI
Ferdia Sherry ( University of Cambridge )
Structural adaptation for noise reduction in magnetic resonance imaging
Karsten Tabelow ( Weiestrass Institute Berlin )
Michael Hintermüller ( Humboldt University and Weierstrass Institute Berlin )
Kostas Papafitsoros ( Weierstrass Institute Berlin )
computed tomography, deep learning, image reconstruction, image segmentation, inverse problems, machine learning, statistical inverse estimation methods