FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386098290> ?p ?o ?g. }

• W4386098290 endingPage "4131" @default.
• W4386098290 startingPage "4131" @default.
• W4386098290 abstract "Hyperspectral image (HSI) super-resolution is a practical and challenging task as it requires the reconstruction of a large number of spectral bands. Achieving excellent reconstruction results can greatly benefit subsequent downstream tasks. The current mainstream hyperspectral super-resolution methods mainly utilize 3D convolutional neural networks (3D CNN) for design. However, the commonly used small kernel size in 3D CNN limits the model’s receptive field, preventing it from considering a wider range of contextual information. Though the receptive field could be expanded by enlarging the kernel size, it results in a dramatic increase in model parameters. Furthermore, the popular vision transformers designed for natural images are not suitable for processing HSI. This is because HSI exhibits sparsity in the spatial domain, which can lead to significant computational resource waste when using self-attention. In this paper, we design a hybrid architecture called HyFormer, which combines the strengths of CNN and transformer for hyperspectral super-resolution. The transformer branch enables intra-spectra interaction to capture fine-grained contextual details at each specific wavelength. Meanwhile, the CNN branch facilitates efficient inter-spectra feature extraction among different wavelengths while maintaining a large receptive field. Specifically, in the transformer branch, we propose a novel Grouping-Aggregation transformer (GAT), comprising grouping self-attention (GSA) and aggregation self-attention (ASA). The GSA is employed to extract diverse fine-grained features of targets, while the ASA facilitates interaction among heterogeneous textures allocated to different channels. In the CNN branch, we propose a Wide-Spanning Separable 3D Attention (WSSA) to enlarge the receptive field while keeping a low parameter number. Building upon WSSA, we construct a wide-spanning CNN module to efficiently extract inter-spectra features. Extensive experiments demonstrate the superior performance of our HyFormer." @default.
• W4386098290 created "2023-08-24" @default.
• W4386098290 creator A5004520838 @default.
• W4386098290 creator A5027316177 @default.
• W4386098290 creator A5034840502 @default.
• W4386098290 creator A5064150384 @default.
• W4386098290 creator A5080268898 @default.
• W4386098290 creator A5081440922 @default.
• W4386098290 date "2023-08-23" @default.
• W4386098290 modified "2023-10-16" @default.
• W4386098290 title "HyFormer: Hybrid Grouping-Aggregation Transformer and Wide-Spanning CNN for Hyperspectral Image Super-Resolution" @default.
• W4386098290 cites W1597864587 @default.
• W4386098290 cites W1885185971 @default.
• W4386098290 cites W1988386267 @default.
• W4386098290 cites W2012946078 @default.
• W4386098290 cites W2055225600 @default.
• W4386098290 cites W2077865152 @default.
• W4386098290 cites W2100109944 @default.
• W4386098290 cites W2129353435 @default.
• W4386098290 cites W2144412886 @default.
• W4386098290 cites W2494274259 @default.
• W4386098290 cites W2761176843 @default.
• W4386098290 cites W2767522909 @default.
• W4386098290 cites W2773850265 @default.
• W4386098290 cites W2790508633 @default.
• W4386098290 cites W2792111852 @default.
• W4386098290 cites W2866634454 @default.
• W4386098290 cites W2883429621 @default.
• W4386098290 cites W2910457605 @default.
• W4386098290 cites W2913036400 @default.
• W4386098290 cites W2955111813 @default.
• W4386098290 cites W3019893222 @default.
• W4386098290 cites W3027120144 @default.
• W4386098290 cites W3044477028 @default.
• W4386098290 cites W3096739052 @default.
• W4386098290 cites W3102745911 @default.
• W4386098290 cites W3124196789 @default.
• W4386098290 cites W3138516171 @default.
• W4386098290 cites W3208181816 @default.
• W4386098290 cites W3208460268 @default.
• W4386098290 cites W4224246728 @default.
• W4386098290 cites W4225991576 @default.
• W4386098290 cites W4285599866 @default.
• W4386098290 cites W4306167924 @default.
• W4386098290 cites W4308235903 @default.
• W4386098290 cites W4312446817 @default.
• W4386098290 cites W4312815172 @default.
• W4386098290 cites W4320901940 @default.
• W4386098290 cites W4382466704 @default.
• W4386098290 cites W4382749525 @default.
• W4386098290 cites W4385525377 @default.
• W4386098290 cites W4386075979 @default.
• W4386098290 doi "https://doi.org/10.3390/rs15174131" @default.
• W4386098290 hasPublicationYear "2023" @default.
• W4386098290 type Work @default.
• W4386098290 citedByCount "0" @default.
• W4386098290 crossrefType "journal-article" @default.
• W4386098290 hasAuthorship W4386098290A5004520838 @default.
• W4386098290 hasAuthorship W4386098290A5027316177 @default.
• W4386098290 hasAuthorship W4386098290A5034840502 @default.
• W4386098290 hasAuthorship W4386098290A5064150384 @default.
• W4386098290 hasAuthorship W4386098290A5080268898 @default.
• W4386098290 hasAuthorship W4386098290A5081440922 @default.
• W4386098290 hasBestOaLocation W43860982901 @default.
• W4386098290 hasConcept C121332964 @default.
• W4386098290 hasConcept C153180895 @default.
• W4386098290 hasConcept C154945302 @default.
• W4386098290 hasConcept C159078339 @default.
• W4386098290 hasConcept C165801399 @default.
• W4386098290 hasConcept C19071747 @default.
• W4386098290 hasConcept C31972630 @default.
• W4386098290 hasConcept C41008148 @default.
• W4386098290 hasConcept C62520636 @default.
• W4386098290 hasConcept C66322947 @default.
• W4386098290 hasConcept C81363708 @default.
• W4386098290 hasConceptScore W4386098290C121332964 @default.
• W4386098290 hasConceptScore W4386098290C153180895 @default.
• W4386098290 hasConceptScore W4386098290C154945302 @default.
• W4386098290 hasConceptScore W4386098290C159078339 @default.
• W4386098290 hasConceptScore W4386098290C165801399 @default.
• W4386098290 hasConceptScore W4386098290C19071747 @default.
• W4386098290 hasConceptScore W4386098290C31972630 @default.
• W4386098290 hasConceptScore W4386098290C41008148 @default.
• W4386098290 hasConceptScore W4386098290C62520636 @default.
• W4386098290 hasConceptScore W4386098290C66322947 @default.
• W4386098290 hasConceptScore W4386098290C81363708 @default.
• W4386098290 hasFunder F4320321001 @default.
• W4386098290 hasIssue "17" @default.
• W4386098290 hasLocation W43860982901 @default.
• W4386098290 hasOpenAccess W4386098290 @default.
• W4386098290 hasPrimaryLocation W43860982901 @default.
• W4386098290 hasRelatedWork W1891287906 @default.
• W4386098290 hasRelatedWork W2028628118 @default.
• W4386098290 hasRelatedWork W2748454020 @default.
• W4386098290 hasRelatedWork W2767651786 @default.
• W4386098290 hasRelatedWork W2781623059 @default.
• W4386098290 hasRelatedWork W2912288872 @default.
• W4386098290 hasRelatedWork W3119610945 @default.

• next