Matches in SemOpenAlex for { <https://semopenalex.org/work/W2946436452> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W2946436452 abstract "Recently, deep neural networks have led to tremendous advances in image super-resolution and achieved remarkable results. However, most deep learning methods attain appreciable performance by increasing model parameters and complexity. Consequently, they are difficult to utilize in common devices despite significant results. With respect to application in resource-limited devices, the goal of the present study involves designing a compact network that exhibits good scalability. We proposed a back projection network (BPnet) as we were inspired by a traditional image super-resolution technique, namely iterative back projection (IBP). Our proposed network is composed of different modules. Each module performs the enlarging function based on the result of the previous module, which is similar to an iteration in IBP. The input of each module network corresponds to the previous down-sampled output minus the low-resolution input image. Additionally, the output of the network is the residual between the ground truth high-resolution image and previous output. The non-linear property of a neural network is maximized through the sparsity of residual input/output. Thus, we can achieve a lightweight network without sacrificing the quality of the results. The experiments indicate that the number of parameters of the proposed BPnet can be less than quarter of those proposed in the state-of-the-art papers and still have comparable results." @default.
- W2946436452 created "2019-05-29" @default.
- W2946436452 creator A5009305504 @default.
- W2946436452 creator A5051147323 @default.
- W2946436452 date "2019-01-01" @default.
- W2946436452 modified "2023-09-27" @default.
- W2946436452 title "Back-Projection Lightweight Network for Accurate Image Super Resolution" @default.
- W2946436452 cites W1677182931 @default.
- W2946436452 cites W1885185971 @default.
- W2946436452 cites W1930824406 @default.
- W2946436452 cites W1950594372 @default.
- W2946436452 cites W2011952414 @default.
- W2946436452 cites W2121058967 @default.
- W2946436452 cites W2194775991 @default.
- W2946436452 cites W2214802144 @default.
- W2946436452 cites W2242218935 @default.
- W2946436452 cites W2263468737 @default.
- W2946436452 cites W2476548250 @default.
- W2946436452 cites W2503339013 @default.
- W2946436452 cites W2508457857 @default.
- W2946436452 cites W2534320940 @default.
- W2946436452 cites W2551161082 @default.
- W2946436452 cites W2557227117 @default.
- W2946436452 cites W2562010997 @default.
- W2946436452 cites W2607041014 @default.
- W2946436452 cites W2740494144 @default.
- W2946436452 cites W2741137940 @default.
- W2946436452 cites W2741196023 @default.
- W2946436452 cites W2747898905 @default.
- W2946436452 cites W2780544323 @default.
- W2946436452 cites W2787779974 @default.
- W2946436452 cites W2795024892 @default.
- W2946436452 cites W2915130236 @default.
- W2946436452 cites W2963372104 @default.
- W2946436452 cites W2963446712 @default.
- W2946436452 cites W2963470893 @default.
- W2946436452 cites W2963729050 @default.
- W2946436452 cites W2964101377 @default.
- W2946436452 cites W2964125708 @default.
- W2946436452 cites W935139217 @default.
- W2946436452 doi "https://doi.org/10.1007/978-3-030-20873-8_9" @default.
- W2946436452 hasPublicationYear "2019" @default.
- W2946436452 type Work @default.
- W2946436452 sameAs 2946436452 @default.
- W2946436452 citedByCount "0" @default.
- W2946436452 crossrefType "book-chapter" @default.
- W2946436452 hasAuthorship W2946436452A5009305504 @default.
- W2946436452 hasAuthorship W2946436452A5051147323 @default.
- W2946436452 hasConcept C11413529 @default.
- W2946436452 hasConcept C115961682 @default.
- W2946436452 hasConcept C138268822 @default.
- W2946436452 hasConcept C146849305 @default.
- W2946436452 hasConcept C154945302 @default.
- W2946436452 hasConcept C155512373 @default.
- W2946436452 hasConcept C205372480 @default.
- W2946436452 hasConcept C41008148 @default.
- W2946436452 hasConcept C48044578 @default.
- W2946436452 hasConcept C50644808 @default.
- W2946436452 hasConcept C57493831 @default.
- W2946436452 hasConcept C77088390 @default.
- W2946436452 hasConceptScore W2946436452C11413529 @default.
- W2946436452 hasConceptScore W2946436452C115961682 @default.
- W2946436452 hasConceptScore W2946436452C138268822 @default.
- W2946436452 hasConceptScore W2946436452C146849305 @default.
- W2946436452 hasConceptScore W2946436452C154945302 @default.
- W2946436452 hasConceptScore W2946436452C155512373 @default.
- W2946436452 hasConceptScore W2946436452C205372480 @default.
- W2946436452 hasConceptScore W2946436452C41008148 @default.
- W2946436452 hasConceptScore W2946436452C48044578 @default.
- W2946436452 hasConceptScore W2946436452C50644808 @default.
- W2946436452 hasConceptScore W2946436452C57493831 @default.
- W2946436452 hasConceptScore W2946436452C77088390 @default.
- W2946436452 hasLocation W29464364521 @default.
- W2946436452 hasOpenAccess W2946436452 @default.
- W2946436452 hasPrimaryLocation W29464364521 @default.
- W2946436452 hasRelatedWork W2783377890 @default.
- W2946436452 hasRelatedWork W2884371179 @default.
- W2946436452 hasRelatedWork W2915895610 @default.
- W2946436452 hasRelatedWork W2950016100 @default.
- W2946436452 hasRelatedWork W2967488183 @default.
- W2946436452 hasRelatedWork W2983814643 @default.
- W2946436452 hasRelatedWork W2984776000 @default.
- W2946436452 hasRelatedWork W3005134776 @default.
- W2946436452 hasRelatedWork W3010092235 @default.
- W2946436452 hasRelatedWork W3015222764 @default.
- W2946436452 hasRelatedWork W3033694261 @default.
- W2946436452 hasRelatedWork W3071046106 @default.
- W2946436452 hasRelatedWork W3087455782 @default.
- W2946436452 hasRelatedWork W3089776790 @default.
- W2946436452 hasRelatedWork W3098619413 @default.
- W2946436452 hasRelatedWork W3128562327 @default.
- W2946436452 hasRelatedWork W3154740884 @default.
- W2946436452 hasRelatedWork W3199488984 @default.
- W2946436452 hasRelatedWork W3204989974 @default.
- W2946436452 hasRelatedWork W3210256214 @default.
- W2946436452 isParatext "false" @default.
- W2946436452 isRetracted "false" @default.
- W2946436452 magId "2946436452" @default.
- W2946436452 workType "book-chapter" @default.