Matches in SemOpenAlex for { <https://semopenalex.org/work/W2936313120> ?p ?o ?g. }
- W2936313120 abstract "Although deep convolutional neural networks (CNNs) have achieved great success in computer vision tasks, its real-world application is still impeded by its voracious demand of computational resources. Current works mostly seek to compress the network by reducing its parameters or parameter-incurred computation, neglecting the influence of the input image on the system complexity. Based on the fact that input images of a CNN contain substantial redundancy, in this paper, we propose a unified framework, dubbed as ThumbNet, to simultaneously accelerate and compress CNN models by enabling them to infer on one thumbnail image. We provide three effective strategies to train ThumbNet. In doing so, ThumbNet learns an inference network that performs equally well on small images as the original-input network on large images. With ThumbNet, not only do we obtain the thumbnail-input inference network that can drastically reduce computation and memory requirements, but also we obtain an image downscaler that can generate thumbnail images for generic classification tasks. Extensive experiments show the effectiveness of ThumbNet, and demonstrate that the thumbnail-input inference network learned by ThumbNet can adequately retain the accuracy of the original-input network even when the input images are downscaled 16 times." @default.
- W2936313120 created "2019-04-25" @default.
- W2936313120 creator A5024763828 @default.
- W2936313120 creator A5086960659 @default.
- W2936313120 date "2019-04-10" @default.
- W2936313120 modified "2023-09-25" @default.
- W2936313120 title "ThumbNet: One Thumbnail Image Contains All You Need for Recognition" @default.
- W2936313120 cites W1487641199 @default.
- W2936313120 cites W1576445103 @default.
- W2936313120 cites W1665214252 @default.
- W2936313120 cites W1690739335 @default.
- W2936313120 cites W1821462560 @default.
- W2936313120 cites W1836465849 @default.
- W2936313120 cites W1922123711 @default.
- W2936313120 cites W1932847118 @default.
- W2936313120 cites W1969430823 @default.
- W2936313120 cites W2002969407 @default.
- W2936313120 cites W2017257315 @default.
- W2936313120 cites W2049819383 @default.
- W2936313120 cites W2091962418 @default.
- W2936313120 cites W2100495367 @default.
- W2936313120 cites W2104636679 @default.
- W2936313120 cites W2108598243 @default.
- W2936313120 cites W2113524221 @default.
- W2936313120 cites W2114766824 @default.
- W2936313120 cites W2131524184 @default.
- W2936313120 cites W2136655611 @default.
- W2936313120 cites W2138857742 @default.
- W2936313120 cites W2167215970 @default.
- W2936313120 cites W2172654076 @default.
- W2936313120 cites W2194775991 @default.
- W2936313120 cites W2233116163 @default.
- W2936313120 cites W2293078015 @default.
- W2936313120 cites W2300242332 @default.
- W2936313120 cites W2319920447 @default.
- W2936313120 cites W2342838199 @default.
- W2936313120 cites W2402144811 @default.
- W2936313120 cites W2419597278 @default.
- W2936313120 cites W2515385951 @default.
- W2936313120 cites W2549139847 @default.
- W2936313120 cites W2566752166 @default.
- W2936313120 cites W2582160025 @default.
- W2936313120 cites W2732026016 @default.
- W2936313120 cites W2750784772 @default.
- W2936313120 cites W2778053709 @default.
- W2936313120 cites W2801201579 @default.
- W2936313120 cites W2808168148 @default.
- W2936313120 cites W2891967877 @default.
- W2936313120 cites W2900808826 @default.
- W2936313120 cites W2941979311 @default.
- W2936313120 cites W2950967261 @default.
- W2936313120 cites W2952008036 @default.
- W2936313120 cites W2962835968 @default.
- W2936313120 cites W2963363373 @default.
- W2936313120 cites W2963382930 @default.
- W2936313120 cites W2963766446 @default.
- W2936313120 cites W2964082701 @default.
- W2936313120 cites W2964233199 @default.
- W2936313120 cites W2964299589 @default.
- W2936313120 doi "https://doi.org/10.48550/arxiv.1904.05034" @default.
- W2936313120 hasPublicationYear "2019" @default.
- W2936313120 type Work @default.
- W2936313120 sameAs 2936313120 @default.
- W2936313120 citedByCount "0" @default.
- W2936313120 crossrefType "posted-content" @default.
- W2936313120 hasAuthorship W2936313120A5024763828 @default.
- W2936313120 hasAuthorship W2936313120A5086960659 @default.
- W2936313120 hasBestOaLocation W29363131201 @default.
- W2936313120 hasConcept C111919701 @default.
- W2936313120 hasConcept C11413529 @default.
- W2936313120 hasConcept C115961682 @default.
- W2936313120 hasConcept C119857082 @default.
- W2936313120 hasConcept C152124472 @default.
- W2936313120 hasConcept C153180895 @default.
- W2936313120 hasConcept C154945302 @default.
- W2936313120 hasConcept C160174412 @default.
- W2936313120 hasConcept C2776214188 @default.
- W2936313120 hasConcept C31972630 @default.
- W2936313120 hasConcept C41008148 @default.
- W2936313120 hasConcept C45374587 @default.
- W2936313120 hasConcept C50644808 @default.
- W2936313120 hasConcept C81363708 @default.
- W2936313120 hasConceptScore W2936313120C111919701 @default.
- W2936313120 hasConceptScore W2936313120C11413529 @default.
- W2936313120 hasConceptScore W2936313120C115961682 @default.
- W2936313120 hasConceptScore W2936313120C119857082 @default.
- W2936313120 hasConceptScore W2936313120C152124472 @default.
- W2936313120 hasConceptScore W2936313120C153180895 @default.
- W2936313120 hasConceptScore W2936313120C154945302 @default.
- W2936313120 hasConceptScore W2936313120C160174412 @default.
- W2936313120 hasConceptScore W2936313120C2776214188 @default.
- W2936313120 hasConceptScore W2936313120C31972630 @default.
- W2936313120 hasConceptScore W2936313120C41008148 @default.
- W2936313120 hasConceptScore W2936313120C45374587 @default.
- W2936313120 hasConceptScore W2936313120C50644808 @default.
- W2936313120 hasConceptScore W2936313120C81363708 @default.
- W2936313120 hasLocation W29363131201 @default.
- W2936313120 hasLocation W29363131202 @default.
- W2936313120 hasOpenAccess W2936313120 @default.
- W2936313120 hasPrimaryLocation W29363131201 @default.