FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2974611604> ?p ?o ?g. }

• W2974611604 abstract "Variational methods are widely applied to ill-posed inverse problems for they have the ability to embed prior knowledge about the solution. However, the level of performance of these methods significantly depends on a set of parameters, which can be estimated through computationally expensive and time-consuming methods. In contrast, deep learning offers very generic and efficient architectures, at the expense of explainability, since it is often used as a black-box, without any fine control over its output. Deep unfolding provides a convenient approach to combine variational-based and deep learning approaches. Starting from a variational formulation for image restoration, we develop iRestNet, a neural network architecture obtained by unfolding a proximal interior point algorithm. Hard constraints, encoding desirable properties for the restored image, are incorporated into the network thanks to a logarithmic barrier, while the barrier parameter, the stepsize, and the penalization weight are learned by the network. We derive explicit expressions for the gradient of the proximity operator for various choices of constraints, which allows training iRestNet with gradient descent and backpropagation. In addition, we provide theoretical results regarding the stability of the network for a common inverse problem example. Numerical experiments on image deblurring problems show that the proposed approach compares favorably with both state-of-the-art variational and machine learning methods in terms of image quality." @default.
• W2974611604 created "2019-09-26" @default.
• W2974611604 creator A5024279313 @default.
• W2974611604 creator A5050720850 @default.
• W2974611604 creator A5060871389 @default.
• W2974611604 creator A5062777204 @default.
• W2974611604 creator A5062845079 @default.
• W2974611604 date "2018-12-11" @default.
• W2974611604 modified "2023-10-15" @default.
• W2974611604 title "Deep Unfolding of a Proximal Interior Point Method for Image Restoration" @default.
• W2974611604 cites W111946688 @default.
• W2974611604 cites W1248559201 @default.
• W2974611604 cites W136351249 @default.
• W2974611604 cites W1457323852 @default.
• W2974611604 cites W1491634526 @default.
• W2974611604 cites W1522301498 @default.
• W2974611604 cites W1580963329 @default.
• W2974611604 cites W1589819097 @default.
• W2974611604 cites W1598281290 @default.
• W2974611604 cites W1831449718 @default.
• W2974611604 cites W1906770428 @default.
• W2974611604 cites W1922442141 @default.
• W2974611604 cites W1946620893 @default.
• W2974611604 cites W1968018625 @default.
• W2974611604 cites W1972868011 @default.
• W2974611604 cites W1973567017 @default.
• W2974611604 cites W1983579205 @default.
• W2974611604 cites W2000200631 @default.
• W2974611604 cites W2006262045 @default.
• W2974611604 cites W2034978228 @default.
• W2974611604 cites W2043781359 @default.
• W2974611604 cites W2052387674 @default.
• W2974611604 cites W205960364 @default.
• W2974611604 cites W2076063813 @default.
• W2974611604 cites W2108763700 @default.
• W2974611604 cites W2114424556 @default.
• W2974611604 cites W2115755118 @default.
• W2974611604 cites W2116901131 @default.
• W2974611604 cites W2118103795 @default.
• W2974611604 cites W2124964692 @default.
• W2974611604 cites W2130975789 @default.
• W2974611604 cites W2133665775 @default.
• W2974611604 cites W2138204001 @default.
• W2974611604 cites W2141538908 @default.
• W2974611604 cites W2142308565 @default.
• W2974611604 cites W2157590940 @default.
• W2974611604 cites W2164278908 @default.
• W2974611604 cites W2168148636 @default.
• W2974611604 cites W2172275395 @default.
• W2974611604 cites W2194775991 @default.
• W2974611604 cites W2228119896 @default.
• W2974611604 cites W2337998167 @default.
• W2974611604 cites W2409744450 @default.
• W2974611604 cites W2496276084 @default.
• W2974611604 cites W2508457857 @default.
• W2974611604 cites W251782979 @default.
• W2974611604 cites W2552111036 @default.
• W2974611604 cites W2552808051 @default.
• W2974611604 cites W2553718801 @default.
• W2974611604 cites W2557283755 @default.
• W2974611604 cites W2574952845 @default.
• W2974611604 cites W2592696501 @default.
• W2974611604 cites W2593624191 @default.
• W2974611604 cites W2613155248 @default.
• W2974611604 cites W2620324432 @default.
• W2974611604 cites W2726656315 @default.
• W2974611604 cites W2770205545 @default.
• W2974611604 cites W2798559986 @default.
• W2974611604 cites W2800894561 @default.
• W2974611604 cites W2889523351 @default.
• W2974611604 cites W2899771611 @default.
• W2974611604 cites W2963470893 @default.
• W2974611604 cites W2963775850 @default.
• W2974611604 cites W2963970238 @default.
• W2974611604 cites W2964153729 @default.
• W2974611604 cites W2980354672 @default.
• W2974611604 cites W3006592723 @default.
• W2974611604 cites W3098900881 @default.
• W2974611604 cites W3100774664 @default.
• W2974611604 cites W3102025760 @default.
• W2974611604 cites W3105465707 @default.
• W2974611604 cites W3125923133 @default.
• W2974611604 cites W3135060683 @default.
• W2974611604 hasPublicationYear "2018" @default.
• W2974611604 type Work @default.
• W2974611604 sameAs 2974611604 @default.
• W2974611604 citedByCount "2" @default.
• W2974611604 countsByYear W29746116042019 @default.
• W2974611604 countsByYear W29746116042021 @default.
• W2974611604 crossrefType "posted-content" @default.
• W2974611604 hasAuthorship W2974611604A5024279313 @default.
• W2974611604 hasAuthorship W2974611604A5050720850 @default.
• W2974611604 hasAuthorship W2974611604A5060871389 @default.
• W2974611604 hasAuthorship W2974611604A5062777204 @default.
• W2974611604 hasAuthorship W2974611604A5062845079 @default.
• W2974611604 hasConcept C104317684 @default.
• W2974611604 hasConcept C106430172 @default.
• W2974611604 hasConcept C108583219 @default.
• W2974611604 hasConcept C112972136 @default.
• W2974611604 hasConcept C11413529 @default.

• next