Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366446336> ?p ?o ?g. }
- W4366446336 endingPage "064006" @default.
- W4366446336 startingPage "064006" @default.
- W4366446336 abstract "Learning neural networks using only few available information is an important ongoing research topic with tremendous potential for applications. In this paper, we introduce a powerful regularizer for the variational modeling of inverse problems in imaging. Our regularizer, called patch normalizing flow regularizer (patchNR), involves a normalizing flow learned on small patches of very few images. In particular, the training is independent of the considered inverse problem such that the same regularizer can be applied for different forward operators acting on the same class of images. By investigating the distribution of patches versus those of the whole image class, we prove that our model is indeed a MAP approach. Numerical examples for low-dose and limited-angle computed tomography (CT) as well as superresolution of material images demonstrate that our method provides very high quality results. The training set consists of just six images for CT and one image for superresolution. Finally, we combine our patchNR with ideas from internal learning for performing superresolution of natural images directly from the low-resolution observation without knowledge of any high-resolution image." @default.
- W4366446336 created "2023-04-22" @default.
- W4366446336 creator A5007073244 @default.
- W4366446336 creator A5008755474 @default.
- W4366446336 creator A5014074603 @default.
- W4366446336 creator A5016133913 @default.
- W4366446336 creator A5026709984 @default.
- W4366446336 creator A5054180832 @default.
- W4366446336 date "2023-05-16" @default.
- W4366446336 modified "2023-10-01" @default.
- W4366446336 title "PatchNR: Learning from very few images by patch normalizing flow regularization" @default.
- W4366446336 cites W1965745576 @default.
- W4366446336 cites W1986649315 @default.
- W4366446336 cites W2004671295 @default.
- W4366446336 cites W2058005980 @default.
- W4366446336 cites W2089441731 @default.
- W4366446336 cites W2103559027 @default.
- W4366446336 cites W2157494358 @default.
- W4366446336 cites W2473963253 @default.
- W4366446336 cites W2573726823 @default.
- W4366446336 cites W2574952845 @default.
- W4366446336 cites W2616747498 @default.
- W4366446336 cites W2730550322 @default.
- W4366446336 cites W2766142070 @default.
- W4366446336 cites W2952020389 @default.
- W4366446336 cites W2963562194 @default.
- W4366446336 cites W3021094251 @default.
- W4366446336 cites W3035304632 @default.
- W4366446336 cites W3040793855 @default.
- W4366446336 cites W3095495550 @default.
- W4366446336 cites W3103586216 @default.
- W4366446336 cites W3105425607 @default.
- W4366446336 cites W3107096356 @default.
- W4366446336 cites W3129907242 @default.
- W4366446336 cites W3134238839 @default.
- W4366446336 cites W3155140410 @default.
- W4366446336 cites W3158187990 @default.
- W4366446336 cites W3172797549 @default.
- W4366446336 cites W3172949370 @default.
- W4366446336 cites W3177141244 @default.
- W4366446336 cites W3180322328 @default.
- W4366446336 cites W3196943155 @default.
- W4366446336 cites W3198284549 @default.
- W4366446336 cites W3211066356 @default.
- W4366446336 cites W3211148078 @default.
- W4366446336 cites W4205273243 @default.
- W4366446336 cites W4211114670 @default.
- W4366446336 doi "https://doi.org/10.1088/1361-6420/acce5e" @default.
- W4366446336 hasPublicationYear "2023" @default.
- W4366446336 type Work @default.
- W4366446336 citedByCount "0" @default.
- W4366446336 crossrefType "journal-article" @default.
- W4366446336 hasAuthorship W4366446336A5007073244 @default.
- W4366446336 hasAuthorship W4366446336A5008755474 @default.
- W4366446336 hasAuthorship W4366446336A5014074603 @default.
- W4366446336 hasAuthorship W4366446336A5016133913 @default.
- W4366446336 hasAuthorship W4366446336A5026709984 @default.
- W4366446336 hasAuthorship W4366446336A5054180832 @default.
- W4366446336 hasBestOaLocation W43664463361 @default.
- W4366446336 hasConcept C11413529 @default.
- W4366446336 hasConcept C115961682 @default.
- W4366446336 hasConcept C134306372 @default.
- W4366446336 hasConcept C135252773 @default.
- W4366446336 hasConcept C141239990 @default.
- W4366446336 hasConcept C153180895 @default.
- W4366446336 hasConcept C154945302 @default.
- W4366446336 hasConcept C207467116 @default.
- W4366446336 hasConcept C2524010 @default.
- W4366446336 hasConcept C2776135515 @default.
- W4366446336 hasConcept C2777212361 @default.
- W4366446336 hasConcept C33923547 @default.
- W4366446336 hasConcept C38349280 @default.
- W4366446336 hasConcept C41008148 @default.
- W4366446336 hasConcept C50644808 @default.
- W4366446336 hasConceptScore W4366446336C11413529 @default.
- W4366446336 hasConceptScore W4366446336C115961682 @default.
- W4366446336 hasConceptScore W4366446336C134306372 @default.
- W4366446336 hasConceptScore W4366446336C135252773 @default.
- W4366446336 hasConceptScore W4366446336C141239990 @default.
- W4366446336 hasConceptScore W4366446336C153180895 @default.
- W4366446336 hasConceptScore W4366446336C154945302 @default.
- W4366446336 hasConceptScore W4366446336C207467116 @default.
- W4366446336 hasConceptScore W4366446336C2524010 @default.
- W4366446336 hasConceptScore W4366446336C2776135515 @default.
- W4366446336 hasConceptScore W4366446336C2777212361 @default.
- W4366446336 hasConceptScore W4366446336C33923547 @default.
- W4366446336 hasConceptScore W4366446336C38349280 @default.
- W4366446336 hasConceptScore W4366446336C41008148 @default.
- W4366446336 hasConceptScore W4366446336C50644808 @default.
- W4366446336 hasFunder F4320320879 @default.
- W4366446336 hasFunder F4320324316 @default.
- W4366446336 hasFunder F4320330291 @default.
- W4366446336 hasIssue "6" @default.
- W4366446336 hasLocation W43664463361 @default.
- W4366446336 hasLocation W43664463362 @default.
- W4366446336 hasOpenAccess W4366446336 @default.
- W4366446336 hasPrimaryLocation W43664463361 @default.
- W4366446336 hasRelatedWork W1761505009 @default.