Matches in SemOpenAlex for { <https://semopenalex.org/work/W2896786907> ?p ?o ?g. }
- W2896786907 endingPage "104" @default.
- W2896786907 startingPage "89" @default.
- W2896786907 abstract "We propose a novel discrete Fourier transform-based pooling layer for convolutional neural networks. The DFT magnitude pooling replaces the traditional max/average pooling layer between the convolution and fully-connected layers to retain translation invariance and shape preserving (aware of shape difference) properties based on the shift theorem of the Fourier transform. Thanks to the ability to handle image misalignment while keeping important structural information in the pooling stage, the DFT magnitude pooling improves the classification accuracy significantly. In addition, we propose the DFT+ method for ensemble networks using the middle convolution layer outputs. The proposed methods are extensively evaluated on various classification tasks using the ImageNet, CUB 2010-2011, MIT Indoors, Caltech 101, FMD and DTD datasets. The AlexNet, VGG-VD 16, Inception-v3, and ResNet are used as the base networks, upon which DFT and DFT+ methods are implemented. Experimental results show that the proposed methods improve the classification performance in all networks and datasets." @default.
- W2896786907 created "2018-10-26" @default.
- W2896786907 creator A5019934281 @default.
- W2896786907 creator A5038404116 @default.
- W2896786907 creator A5061007351 @default.
- W2896786907 date "2018-01-01" @default.
- W2896786907 modified "2023-09-23" @default.
- W2896786907 title "DFT-based Transformation Invariant Pooling Layer for Visual Classification" @default.
- W2896786907 cites W1524680991 @default.
- W2896786907 cites W1536680647 @default.
- W2896786907 cites W1898560071 @default.
- W2896786907 cites W1903029394 @default.
- W2896786907 cites W1909952827 @default.
- W2896786907 cites W1948751323 @default.
- W2896786907 cites W2047643928 @default.
- W2896786907 cites W2059552065 @default.
- W2896786907 cites W2097117768 @default.
- W2896786907 cites W2102605133 @default.
- W2896786907 cites W2104657103 @default.
- W2896786907 cites W2117539524 @default.
- W2896786907 cites W2147238549 @default.
- W2896786907 cites W2152161678 @default.
- W2896786907 cites W2166049352 @default.
- W2896786907 cites W2183341477 @default.
- W2896786907 cites W2194775991 @default.
- W2896786907 cites W2204257188 @default.
- W2896786907 cites W2479109623 @default.
- W2896786907 cites W2566365295 @default.
- W2896786907 cites W2740620254 @default.
- W2896786907 cites W2963037989 @default.
- W2896786907 cites W2963066927 @default.
- W2896786907 cites W2963108253 @default.
- W2896786907 cites W2963239066 @default.
- W2896786907 cites W2964275061 @default.
- W2896786907 cites W3104752576 @default.
- W2896786907 doi "https://doi.org/10.1007/978-3-030-01264-9_6" @default.
- W2896786907 hasPublicationYear "2018" @default.
- W2896786907 type Work @default.
- W2896786907 sameAs 2896786907 @default.
- W2896786907 citedByCount "24" @default.
- W2896786907 countsByYear W28967869072018 @default.
- W2896786907 countsByYear W28967869072019 @default.
- W2896786907 countsByYear W28967869072020 @default.
- W2896786907 countsByYear W28967869072021 @default.
- W2896786907 countsByYear W28967869072022 @default.
- W2896786907 countsByYear W28967869072023 @default.
- W2896786907 crossrefType "book-chapter" @default.
- W2896786907 hasAuthorship W2896786907A5019934281 @default.
- W2896786907 hasAuthorship W2896786907A5038404116 @default.
- W2896786907 hasAuthorship W2896786907A5061007351 @default.
- W2896786907 hasConcept C102519508 @default.
- W2896786907 hasConcept C104317684 @default.
- W2896786907 hasConcept C11413529 @default.
- W2896786907 hasConcept C134306372 @default.
- W2896786907 hasConcept C153180895 @default.
- W2896786907 hasConcept C154945302 @default.
- W2896786907 hasConcept C185592680 @default.
- W2896786907 hasConcept C190470478 @default.
- W2896786907 hasConcept C203024314 @default.
- W2896786907 hasConcept C204241405 @default.
- W2896786907 hasConcept C33923547 @default.
- W2896786907 hasConcept C37914503 @default.
- W2896786907 hasConcept C41008148 @default.
- W2896786907 hasConcept C45347329 @default.
- W2896786907 hasConcept C50644808 @default.
- W2896786907 hasConcept C55493867 @default.
- W2896786907 hasConcept C57733114 @default.
- W2896786907 hasConcept C70437156 @default.
- W2896786907 hasConcept C76563020 @default.
- W2896786907 hasConcept C81363708 @default.
- W2896786907 hasConceptScore W2896786907C102519508 @default.
- W2896786907 hasConceptScore W2896786907C104317684 @default.
- W2896786907 hasConceptScore W2896786907C11413529 @default.
- W2896786907 hasConceptScore W2896786907C134306372 @default.
- W2896786907 hasConceptScore W2896786907C153180895 @default.
- W2896786907 hasConceptScore W2896786907C154945302 @default.
- W2896786907 hasConceptScore W2896786907C185592680 @default.
- W2896786907 hasConceptScore W2896786907C190470478 @default.
- W2896786907 hasConceptScore W2896786907C203024314 @default.
- W2896786907 hasConceptScore W2896786907C204241405 @default.
- W2896786907 hasConceptScore W2896786907C33923547 @default.
- W2896786907 hasConceptScore W2896786907C37914503 @default.
- W2896786907 hasConceptScore W2896786907C41008148 @default.
- W2896786907 hasConceptScore W2896786907C45347329 @default.
- W2896786907 hasConceptScore W2896786907C50644808 @default.
- W2896786907 hasConceptScore W2896786907C55493867 @default.
- W2896786907 hasConceptScore W2896786907C57733114 @default.
- W2896786907 hasConceptScore W2896786907C70437156 @default.
- W2896786907 hasConceptScore W2896786907C76563020 @default.
- W2896786907 hasConceptScore W2896786907C81363708 @default.
- W2896786907 hasLocation W28967869071 @default.
- W2896786907 hasOpenAccess W2896786907 @default.
- W2896786907 hasPrimaryLocation W28967869071 @default.
- W2896786907 hasRelatedWork W2500293854 @default.
- W2896786907 hasRelatedWork W2514274290 @default.
- W2896786907 hasRelatedWork W2517027266 @default.
- W2896786907 hasRelatedWork W2758063741 @default.
- W2896786907 hasRelatedWork W2792080776 @default.