BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Learning better models for inverse problems in imaging with an app
 lication to demosaicing - Joana Grah\, TU Graz
DTSTART:20181119T160000Z
DTEND:20181119T170000Z
UID:TALK114481@talks.cam.ac.uk
CONTACT:Yury Korolev
DESCRIPTION:In this talk\, we will present our ongoing activities in learn
 ing better models for inverse problems in imaging. We consider classical v
 ariational models used for inverse problems but generalise these models by
  introducing a large number of free model parameters. We learn the free mo
 del parameters by minimising a loss function comparing the reconstructed i
 mages obtained from the variational models with ground truth solutions fro
 m a training data base. We will also show recent results on learning "deep
 er" regularisers that are allowed to change their parameters in each itera
 tion of the algorithm. We show applications to different inverse problems 
 in imaging\, where we put a particular focus on joint image demosaicing an
 d denoising.
LOCATION:MR 5\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR