Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385488928> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4385488928 abstract "In this paper, we propose an Effective and lightweight Transformer for image shadow detection and removal named SpA-Former to recover a shadow-free image from a single shaded image. In contrast to conventional methods that require two stages for shadow detection and then shadow removal, the SpA-Former is a one-stage network capable of learning the mapping function between shadows and no shadows, and does not require a separate shadow detection. SpA-Former is composed of Transformer encoder and CNN decoder, where the CNN decoder contains the GAN network. In the Transformer encoding stage, Gated Feed-Forward Network(GFFN) is devised to control the information flow. In the CNN decoding stage, Two-wheel RNN joint spatial attention(TWRNN) and Fourier transform residual block (FTR) are designed to achieve satisfactory results in shadow removal. The combination of Transformer and CNN is able to feed global features from the Vision Transformer encoder into CNN to enhance the global perception of CNN branches, taking into account the complementarity of local features and the global. The SpA-Former's inference speed is 0.0459s, and the final Parameters and FLOPS are only 0.47MB and 15G, achieving the current lightweight of image shadow removal. The source code of MemoryNet can be obtained from https://github.com/zhangbaijin/SpA-Former-shadow-removal" @default.
- W4385488928 created "2023-08-03" @default.
- W4385488928 creator A5007013035 @default.
- W4385488928 creator A5030921410 @default.
- W4385488928 creator A5040397688 @default.
- W4385488928 creator A5063963886 @default.
- W4385488928 creator A5068679143 @default.
- W4385488928 date "2023-06-18" @default.
- W4385488928 modified "2023-09-26" @default.
- W4385488928 title "SpA-Former:An Effective and lightweight Transformer for image shadow removal" @default.
- W4385488928 cites W1967913888 @default.
- W4385488928 cites W2060850257 @default.
- W4385488928 cites W2131579142 @default.
- W4385488928 cites W2740832349 @default.
- W4385488928 cites W2756177297 @default.
- W4385488928 cites W2801375052 @default.
- W4385488928 cites W2900701924 @default.
- W4385488928 cites W2969075423 @default.
- W4385488928 cites W2997085666 @default.
- W4385488928 cites W2997378071 @default.
- W4385488928 cites W2998154526 @default.
- W4385488928 cites W3032151187 @default.
- W4385488928 cites W3104943862 @default.
- W4385488928 cites W3106147826 @default.
- W4385488928 cites W3118389845 @default.
- W4385488928 cites W3176932505 @default.
- W4385488928 cites W3178680498 @default.
- W4385488928 cites W3188140417 @default.
- W4385488928 cites W4200338675 @default.
- W4385488928 cites W4214552968 @default.
- W4385488928 cites W4223607325 @default.
- W4385488928 cites W4224257027 @default.
- W4385488928 cites W4225672218 @default.
- W4385488928 cites W4232877268 @default.
- W4385488928 cites W4283815594 @default.
- W4385488928 cites W4312187100 @default.
- W4385488928 cites W4312549046 @default.
- W4385488928 cites W4312862560 @default.
- W4385488928 cites W4313427650 @default.
- W4385488928 doi "https://doi.org/10.1109/ijcnn54540.2023.10191081" @default.
- W4385488928 hasPublicationYear "2023" @default.
- W4385488928 type Work @default.
- W4385488928 citedByCount "0" @default.
- W4385488928 crossrefType "proceedings-article" @default.
- W4385488928 hasAuthorship W4385488928A5007013035 @default.
- W4385488928 hasAuthorship W4385488928A5030921410 @default.
- W4385488928 hasAuthorship W4385488928A5040397688 @default.
- W4385488928 hasAuthorship W4385488928A5063963886 @default.
- W4385488928 hasAuthorship W4385488928A5068679143 @default.
- W4385488928 hasConcept C111919701 @default.
- W4385488928 hasConcept C11413529 @default.
- W4385488928 hasConcept C118505674 @default.
- W4385488928 hasConcept C119599485 @default.
- W4385488928 hasConcept C127413603 @default.
- W4385488928 hasConcept C154945302 @default.
- W4385488928 hasConcept C155512373 @default.
- W4385488928 hasConcept C165801399 @default.
- W4385488928 hasConcept C31972630 @default.
- W4385488928 hasConcept C41008148 @default.
- W4385488928 hasConcept C57273362 @default.
- W4385488928 hasConcept C66322947 @default.
- W4385488928 hasConcept C81363708 @default.
- W4385488928 hasConceptScore W4385488928C111919701 @default.
- W4385488928 hasConceptScore W4385488928C11413529 @default.
- W4385488928 hasConceptScore W4385488928C118505674 @default.
- W4385488928 hasConceptScore W4385488928C119599485 @default.
- W4385488928 hasConceptScore W4385488928C127413603 @default.
- W4385488928 hasConceptScore W4385488928C154945302 @default.
- W4385488928 hasConceptScore W4385488928C155512373 @default.
- W4385488928 hasConceptScore W4385488928C165801399 @default.
- W4385488928 hasConceptScore W4385488928C31972630 @default.
- W4385488928 hasConceptScore W4385488928C41008148 @default.
- W4385488928 hasConceptScore W4385488928C57273362 @default.
- W4385488928 hasConceptScore W4385488928C66322947 @default.
- W4385488928 hasConceptScore W4385488928C81363708 @default.
- W4385488928 hasFunder F4320321001 @default.
- W4385488928 hasLocation W43854889281 @default.
- W4385488928 hasOpenAccess W4385488928 @default.
- W4385488928 hasPrimaryLocation W43854889281 @default.
- W4385488928 hasRelatedWork W1891287906 @default.
- W4385488928 hasRelatedWork W1969923398 @default.
- W4385488928 hasRelatedWork W2036807459 @default.
- W4385488928 hasRelatedWork W2166024367 @default.
- W4385488928 hasRelatedWork W2275988210 @default.
- W4385488928 hasRelatedWork W2358034992 @default.
- W4385488928 hasRelatedWork W2367936931 @default.
- W4385488928 hasRelatedWork W2772917594 @default.
- W4385488928 hasRelatedWork W2775347418 @default.
- W4385488928 hasRelatedWork W3181746755 @default.
- W4385488928 isParatext "false" @default.
- W4385488928 isRetracted "false" @default.
- W4385488928 workType "article" @default.