FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2936614765> ?p ?o ?g. }

• W2936614765 endingPage "1" @default.
• W2936614765 startingPage "1" @default.
• W2936614765 abstract "Compared with global average pooling in existing deep convolutional neural networks (CNNs), global covariance pooling can capture richer statistics of deep features, having potential for improving representation and generalization abilities of deep CNNs. However, integration of global covariance pooling into deep CNNs brings two challenges: (1) robust covariance estimation given deep features of high dimension and small sample size; (2) appropriate usage of geometry of covariances. To address these challenges, we propose a global Matrix Power Normalized COVariance (MPN-COV) Pooling. Our MPN-COV conforms to a robust covariance estimator, very suitable for scenario of high dimension and small sample size. It can also be regarded as Power-Euclidean metric between covariances, effectively exploiting their geometry. Furthermore, a global Gaussian embedding network is proposed to incorporate first-order statistics into MPN-COV. For fast training of MPN-COV networks, we implement an iterative matrix square root normalization, avoiding GPU unfriendly eigen-decomposition inherent in MPN-COV. Additionally, progressive 1x1 convolutions and group convolution are introduced to compress covariance representations. The proposed methods are highly modular, readily plugged into existing deep CNNs. Extensive experiments are conducted on large-scale object classification, scene categorization, fine-grained visual recognition and texture classification, showing our methods outperform the counterparts and obtain state-of-the-art performance." @default.
• W2936614765 created "2019-04-25" @default.
• W2936614765 creator A5018318136 @default.
• W2936614765 creator A5029031269 @default.
• W2936614765 creator A5052272812 @default.
• W2936614765 creator A5070918242 @default.
• W2936614765 creator A5086215037 @default.
• W2936614765 date "2020-01-01" @default.
• W2936614765 modified "2023-10-17" @default.
• W2936614765 title "Deep CNNs Meet Global Covariance Pooling: Better Representation and Generalization" @default.
• W2936614765 cites W1483804921 @default.
• W2936614765 cites W1549083695 @default.
• W2936614765 cites W1598557217 @default.
• W2936614765 cites W1677182931 @default.
• W2936614765 cites W1963882359 @default.
• W2936614765 cites W1966385142 @default.
• W2936614765 cites W1983496390 @default.
• W2936614765 cites W1984309565 @default.
• W2936614765 cites W1996955012 @default.
• W2936614765 cites W2006766164 @default.
• W2936614765 cites W2024165284 @default.
• W2936614765 cites W2029440543 @default.
• W2936614765 cites W2037900003 @default.
• W2936614765 cites W2047643928 @default.
• W2936614765 cites W2062125287 @default.
• W2936614765 cites W2097117768 @default.
• W2936614765 cites W2104657103 @default.
• W2936614765 cites W2106139345 @default.
• W2936614765 cites W2108598243 @default.
• W2936614765 cites W2112759033 @default.
• W2936614765 cites W2119138423 @default.
• W2936614765 cites W2119963163 @default.
• W2936614765 cites W2138011018 @default.
• W2936614765 cites W2151103935 @default.
• W2936614765 cites W2152161678 @default.
• W2936614765 cites W2163922914 @default.
• W2936614765 cites W2194775991 @default.
• W2936614765 cites W2204257188 @default.
• W2936614765 cites W2207669591 @default.
• W2936614765 cites W2324076434 @default.
• W2936614765 cites W2343076010 @default.
• W2936614765 cites W2395611524 @default.
• W2936614765 cites W2423822072 @default.
• W2936614765 cites W2549139847 @default.
• W2936614765 cites W2551829638 @default.
• W2936614765 cites W2554320282 @default.
• W2936614765 cites W2566365295 @default.
• W2936614765 cites W2601789736 @default.
• W2936614765 cites W2604414494 @default.
• W2936614765 cites W2731821979 @default.
• W2936614765 cites W2732026016 @default.
• W2936614765 cites W2740620254 @default.
• W2936614765 cites W2749507674 @default.
• W2936614765 cites W2752782242 @default.
• W2936614765 cites W2780534582 @default.
• W2936614765 cites W2780838211 @default.
• W2936614765 cites W2900081457 @default.
• W2936614765 cites W2951019013 @default.
• W2936614765 cites W2962761264 @default.
• W2936614765 cites W2963066927 @default.
• W2936614765 cites W2963173190 @default.
• W2936614765 cites W2963363102 @default.
• W2936614765 cites W2963446712 @default.
• W2936614765 cites W2963550079 @default.
• W2936614765 cites W2964044082 @default.
• W2936614765 cites W2964350570 @default.
• W2936614765 cites W639708223 @default.
• W2936614765 doi "https://doi.org/10.1109/tpami.2020.2974833" @default.
• W2936614765 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/32086198" @default.
• W2936614765 hasPublicationYear "2020" @default.
• W2936614765 type Work @default.
• W2936614765 sameAs 2936614765 @default.
• W2936614765 citedByCount "35" @default.
• W2936614765 countsByYear W29366147652020 @default.
• W2936614765 countsByYear W29366147652021 @default.
• W2936614765 countsByYear W29366147652022 @default.
• W2936614765 countsByYear W29366147652023 @default.
• W2936614765 crossrefType "journal-article" @default.
• W2936614765 hasAuthorship W2936614765A5018318136 @default.
• W2936614765 hasAuthorship W2936614765A5029031269 @default.
• W2936614765 hasAuthorship W2936614765A5052272812 @default.
• W2936614765 hasAuthorship W2936614765A5070918242 @default.
• W2936614765 hasAuthorship W2936614765A5086215037 @default.
• W2936614765 hasBestOaLocation W29366147652 @default.
• W2936614765 hasConcept C105795698 @default.
• W2936614765 hasConcept C108583219 @default.
• W2936614765 hasConcept C11413529 @default.
• W2936614765 hasConcept C118006245 @default.
• W2936614765 hasConcept C153180895 @default.
• W2936614765 hasConcept C154945302 @default.
• W2936614765 hasConcept C162324750 @default.
• W2936614765 hasConcept C176217482 @default.
• W2936614765 hasConcept C178650346 @default.
• W2936614765 hasConcept C180877172 @default.
• W2936614765 hasConcept C185142706 @default.
• W2936614765 hasConcept C21547014 @default.
• W2936614765 hasConcept C33923547 @default.
• W2936614765 hasConcept C41008148 @default.

• next