MS631 Optimization methods in imaging and learning: From continuous to discrete and reverse

undefinedN. Thorstensen, O. Scherzer
Universitaet Wien/AT

 

Optimization methods in imaging and learning: From continuous to discrete and reverse

In this minisymposium we aim for bringing together researchers from different communities, from learning and optimization, which investigate the problem from discrete and continuous points of view, respectively.

The topics addressed in this minisymposium are:

  1. Modeling of suitable priors in the continuous and discrete setting, hybrid approaches.
  2. Mathematical analysis (convex vs. non-convex, error bounds, well-posedness).
  3. Numerical solvers (recent techniques, trends in numerical/combinatorial optimization).
  4. Discrete and continuous network flows for energy minmization

The goal of this mini symposium is to communicate recent progress in discrete and continuous optimization methods for imaging and vision. Solving continuous PDE models using discrete optimization methods such as graph cut techniques. Asymptotic, understanding the connection between continuous solution and discrete approximation – for instance, which energies are graph representable.

 

 

>> submit an abstract online

Print Version