Discriminative Transfer Learning for General Image Restoration

Lei Xiao, Felix Heide, Wolfgang Heidrich, Bernhard Schölkopf, Michael Hirsch

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

Recently, several discriminative learning approaches have been proposed for effective image restoration, achieving convincing trade-off between image quality and computational efficiency. However, these methods require separate training for each restoration task (e.g., denoising, deblurring, demosaicing) and problem condition (e.g., noise level of input images). This makes it time-consuming and difficult to encompass all tasks and conditions during training. In this paper, we propose a discriminative transfer learning method that incorporates formal proximal optimization and discriminative learning for general image restoration. The method requires a single-pass discriminative training and allows for reuse across various problems and conditions while achieving an efficiency comparable to previous discriminative approaches. Furthermore, after being trained, our model can be easily transferred to new likelihood terms to solve untrained tasks, or be combined with existing priors to further improve image restoration quality.
Original languageEnglish (US)
Pages (from-to)4091-4104
Number of pages14
JournalIEEE Transactions on Image Processing
Volume27
Issue number8
DOIs
StatePublished - Apr 30 2018

Fingerprint Dive into the research topics of 'Discriminative Transfer Learning for General Image Restoration'. Together they form a unique fingerprint.

Cite this