Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287239779> ?p ?o ?g. }
- W4287239779 endingPage "681" @default.
- W4287239779 startingPage "668" @default.
- W4287239779 abstract "Unlike the success of neural architecture search (NAS) in high-level vision tasks, it remains challenging to find computationally efficient and memory-efficient solutions to low-level vision problems such as image restoration through NAS. One of the fundamental barriers to differential NAS-based image restoration is the optimization gap between the super-network and the sub-architectures, causing instability during the searching process. In this paper, we present a novel approach to fill this gap in image denoising application by connecting model-guided design (MoD) with NAS (MoD-NAS). Specifically, we propose to construct a new search space under a model-guided framework and develop more stable and efficient differential search strategies. MoD-NAS employs a highly reusable width search strategy and a densely connected search block to automatically select the operations of each layer as well as network width and depth via gradient descent. During the search process, the proposed MoD-NAS remains stable because of the smoother search space designed under the model-guided framework. Experimental results on several popular datasets show that our MoD-NAS method has achieved at least comparable even better PSNR performance than current state-of-the-art methods with fewer parameters, fewer flops, and less testing time. “The code associate with this paper is available at: <uri xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>https://see.xidian.edu.cn/faculty/wsdong/Projects/Mod-NAS.htm</uri> ”." @default.
- W4287239779 created "2022-07-25" @default.
- W4287239779 creator A5022200044 @default.
- W4287239779 creator A5022526821 @default.
- W4287239779 creator A5033615240 @default.
- W4287239779 creator A5037310802 @default.
- W4287239779 creator A5074780876 @default.
- W4287239779 creator A5077166272 @default.
- W4287239779 date "2023-01-01" @default.
- W4287239779 modified "2023-10-16" @default.
- W4287239779 title "Searching Efficient Model-Guided Deep Network for Image Denoising" @default.
- W4287239779 cites W1588663000 @default.
- W4287239779 cites W1906770428 @default.
- W4287239779 cites W1930824406 @default.
- W4287239779 cites W2056370875 @default.
- W4287239779 cites W2121927366 @default.
- W4287239779 cites W2131024476 @default.
- W4287239779 cites W2149669120 @default.
- W4287239779 cites W2556068545 @default.
- W4287239779 cites W2613155248 @default.
- W4287239779 cites W2784344583 @default.
- W4287239779 cites W2799192307 @default.
- W4287239779 cites W2948667002 @default.
- W4287239779 cites W2962935103 @default.
- W4287239779 cites W2963085671 @default.
- W4287239779 cites W2963136578 @default.
- W4287239779 cites W2963163009 @default.
- W4287239779 cites W2963372104 @default.
- W4287239779 cites W2963725279 @default.
- W4287239779 cites W2963774720 @default.
- W4287239779 cites W2963918968 @default.
- W4287239779 cites W2964081807 @default.
- W4287239779 cites W2964101377 @default.
- W4287239779 cites W2964125708 @default.
- W4287239779 cites W2965566740 @default.
- W4287239779 cites W2965658867 @default.
- W4287239779 cites W2967733054 @default.
- W4287239779 cites W2979122372 @default.
- W4287239779 cites W2997104867 @default.
- W4287239779 cites W3000775737 @default.
- W4287239779 cites W3034278302 @default.
- W4287239779 cites W3034516944 @default.
- W4287239779 cites W3034652926 @default.
- W4287239779 cites W3035122910 @default.
- W4287239779 cites W3035223629 @default.
- W4287239779 cites W3035302306 @default.
- W4287239779 cites W3035383808 @default.
- W4287239779 cites W3085109118 @default.
- W4287239779 cites W3099686304 @default.
- W4287239779 cites W3105260381 @default.
- W4287239779 cites W3169761664 @default.
- W4287239779 cites W3173217100 @default.
- W4287239779 cites W4242059867 @default.
- W4287239779 cites W4285606127 @default.
- W4287239779 doi "https://doi.org/10.1109/tip.2022.3231741" @default.
- W4287239779 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37015619" @default.
- W4287239779 hasPublicationYear "2023" @default.
- W4287239779 type Work @default.
- W4287239779 citedByCount "1" @default.
- W4287239779 countsByYear W42872397792023 @default.
- W4287239779 crossrefType "journal-article" @default.
- W4287239779 hasAuthorship W4287239779A5022200044 @default.
- W4287239779 hasAuthorship W4287239779A5022526821 @default.
- W4287239779 hasAuthorship W4287239779A5033615240 @default.
- W4287239779 hasAuthorship W4287239779A5037310802 @default.
- W4287239779 hasAuthorship W4287239779A5074780876 @default.
- W4287239779 hasAuthorship W4287239779A5077166272 @default.
- W4287239779 hasBestOaLocation W42872397792 @default.
- W4287239779 hasConcept C111919701 @default.
- W4287239779 hasConcept C11413529 @default.
- W4287239779 hasConcept C115961682 @default.
- W4287239779 hasConcept C127413603 @default.
- W4287239779 hasConcept C146978453 @default.
- W4287239779 hasConcept C153180895 @default.
- W4287239779 hasConcept C154945302 @default.
- W4287239779 hasConcept C177264268 @default.
- W4287239779 hasConcept C199360897 @default.
- W4287239779 hasConcept C2524010 @default.
- W4287239779 hasConcept C2776760102 @default.
- W4287239779 hasConcept C2777210771 @default.
- W4287239779 hasConcept C2780801425 @default.
- W4287239779 hasConcept C33923547 @default.
- W4287239779 hasConcept C41008148 @default.
- W4287239779 hasConcept C93226319 @default.
- W4287239779 hasConcept C98045186 @default.
- W4287239779 hasConceptScore W4287239779C111919701 @default.
- W4287239779 hasConceptScore W4287239779C11413529 @default.
- W4287239779 hasConceptScore W4287239779C115961682 @default.
- W4287239779 hasConceptScore W4287239779C127413603 @default.
- W4287239779 hasConceptScore W4287239779C146978453 @default.
- W4287239779 hasConceptScore W4287239779C153180895 @default.
- W4287239779 hasConceptScore W4287239779C154945302 @default.
- W4287239779 hasConceptScore W4287239779C177264268 @default.
- W4287239779 hasConceptScore W4287239779C199360897 @default.
- W4287239779 hasConceptScore W4287239779C2524010 @default.
- W4287239779 hasConceptScore W4287239779C2776760102 @default.
- W4287239779 hasConceptScore W4287239779C2777210771 @default.
- W4287239779 hasConceptScore W4287239779C2780801425 @default.