Matches in SemOpenAlex for { <https://semopenalex.org/work/W4353047582> ?p ?o ?g. }
- W4353047582 endingPage "110204" @default.
- W4353047582 startingPage "110204" @default.
- W4353047582 abstract "Convolutional Neural Networks (CNN) are widely used in image dehazing. However, existing network frameworks are built based on manual design from practical experience, lacking interpretable result or theoretical guidelines. Recently, residual networks are regarded as the explicit Euler forward approximation of the ODE (Ordinary Differential Equation), and several ODE-inspired networks are proposed based on the low-order explicit Euler schemes. However, on the issues of system stability and training convergence, high-order Implicit Adams Predictor–Corrector (IAPC) methods have proven to be better than low-order explicit Euler methods. Hence, we extend the IAPC method to the High-order Implicit Adams Network (HIAN). To do so, we design a series of Implicit Adams Predictor–Corrector Blocks (IABs) based on the high-order IAPC methods, all of which give better stability and accuracy than the ones designed using the low-order Euler methods. Given that, we further propose the Implicit Adams Predictor–Corrector Module (IAM) by combining the Non-local Sparse Attention (NSA) and Attention Feature Fusion (AFF) with stacked IABs where the NSA explores the mutual-correlation among intermediate features with low computation cost via a sparse constraint, while the AFF fuses intermediate features by reweighting the features from stacked IABs adaptively. Moreover, because manual network design with IABs limits dehazing performance, the Neural Architecture Search (NAS) is used to find an optimal architecture automatically. This resulting design not only is interpretable for image dehazing but also provides a reliable guideline on future network designs. The experiments demonstrate that the proposed method outperforms most existing methods on both synthetic and real images." @default.
- W4353047582 created "2023-03-23" @default.
- W4353047582 creator A5048778510 @default.
- W4353047582 creator A5054391201 @default.
- W4353047582 creator A5081300658 @default.
- W4353047582 creator A5086489308 @default.
- W4353047582 date "2023-05-01" @default.
- W4353047582 modified "2023-09-27" @default.
- W4353047582 title "High-order Adams Network (HIAN) for image dehazing" @default.
- W4353047582 cites W2147318913 @default.
- W4353047582 cites W2467473805 @default.
- W4353047582 cites W2779176852 @default.
- W4353047582 cites W2792830341 @default.
- W4353047582 cites W2947156405 @default.
- W4353047582 cites W2948606054 @default.
- W4353047582 cites W2963152299 @default.
- W4353047582 cites W2963306157 @default.
- W4353047582 cites W2963928582 @default.
- W4353047582 cites W2967733054 @default.
- W4353047582 cites W2969218019 @default.
- W4353047582 cites W2986969740 @default.
- W4353047582 cites W2990007814 @default.
- W4353047582 cites W2997210448 @default.
- W4353047582 cites W2997614040 @default.
- W4353047582 cites W2998249728 @default.
- W4353047582 cites W3004101291 @default.
- W4353047582 cites W3024939336 @default.
- W4353047582 cites W3034331889 @default.
- W4353047582 cites W3034516944 @default.
- W4353047582 cites W3035631962 @default.
- W4353047582 cites W3036625258 @default.
- W4353047582 cites W3105260381 @default.
- W4353047582 cites W3111785512 @default.
- W4353047582 cites W3131333716 @default.
- W4353047582 cites W3174531399 @default.
- W4353047582 cites W3176583247 @default.
- W4353047582 cites W3177127346 @default.
- W4353047582 cites W4200023362 @default.
- W4353047582 cites W4225141093 @default.
- W4353047582 cites W4289986018 @default.
- W4353047582 cites W4294811441 @default.
- W4353047582 doi "https://doi.org/10.1016/j.asoc.2023.110204" @default.
- W4353047582 hasPublicationYear "2023" @default.
- W4353047582 type Work @default.
- W4353047582 citedByCount "0" @default.
- W4353047582 crossrefType "journal-article" @default.
- W4353047582 hasAuthorship W4353047582A5048778510 @default.
- W4353047582 hasAuthorship W4353047582A5054391201 @default.
- W4353047582 hasAuthorship W4353047582A5081300658 @default.
- W4353047582 hasAuthorship W4353047582A5086489308 @default.
- W4353047582 hasConcept C112972136 @default.
- W4353047582 hasConcept C11413529 @default.
- W4353047582 hasConcept C115961682 @default.
- W4353047582 hasConcept C119857082 @default.
- W4353047582 hasConcept C126255220 @default.
- W4353047582 hasConcept C134306372 @default.
- W4353047582 hasConcept C154945302 @default.
- W4353047582 hasConcept C155512373 @default.
- W4353047582 hasConcept C162324750 @default.
- W4353047582 hasConcept C2777303404 @default.
- W4353047582 hasConcept C28826006 @default.
- W4353047582 hasConcept C33923547 @default.
- W4353047582 hasConcept C34862557 @default.
- W4353047582 hasConcept C38409319 @default.
- W4353047582 hasConcept C41008148 @default.
- W4353047582 hasConcept C50522688 @default.
- W4353047582 hasConcept C50644808 @default.
- W4353047582 hasConcept C51544822 @default.
- W4353047582 hasConcept C62884695 @default.
- W4353047582 hasConcept C768646 @default.
- W4353047582 hasConcept C78045399 @default.
- W4353047582 hasConcept C81363708 @default.
- W4353047582 hasConceptScore W4353047582C112972136 @default.
- W4353047582 hasConceptScore W4353047582C11413529 @default.
- W4353047582 hasConceptScore W4353047582C115961682 @default.
- W4353047582 hasConceptScore W4353047582C119857082 @default.
- W4353047582 hasConceptScore W4353047582C126255220 @default.
- W4353047582 hasConceptScore W4353047582C134306372 @default.
- W4353047582 hasConceptScore W4353047582C154945302 @default.
- W4353047582 hasConceptScore W4353047582C155512373 @default.
- W4353047582 hasConceptScore W4353047582C162324750 @default.
- W4353047582 hasConceptScore W4353047582C2777303404 @default.
- W4353047582 hasConceptScore W4353047582C28826006 @default.
- W4353047582 hasConceptScore W4353047582C33923547 @default.
- W4353047582 hasConceptScore W4353047582C34862557 @default.
- W4353047582 hasConceptScore W4353047582C38409319 @default.
- W4353047582 hasConceptScore W4353047582C41008148 @default.
- W4353047582 hasConceptScore W4353047582C50522688 @default.
- W4353047582 hasConceptScore W4353047582C50644808 @default.
- W4353047582 hasConceptScore W4353047582C51544822 @default.
- W4353047582 hasConceptScore W4353047582C62884695 @default.
- W4353047582 hasConceptScore W4353047582C768646 @default.
- W4353047582 hasConceptScore W4353047582C78045399 @default.
- W4353047582 hasConceptScore W4353047582C81363708 @default.
- W4353047582 hasLocation W43530475821 @default.
- W4353047582 hasOpenAccess W4353047582 @default.
- W4353047582 hasPrimaryLocation W43530475821 @default.
- W4353047582 hasRelatedWork W1501782796 @default.