FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2601789736> ?p ?o ?g. }

• W2601789736 abstract "By stacking layers of convolution and nonlinearity, convolutional networks (ConvNets) effectively learn from lowlevel to high-level features and discriminative representations. Since the end goal of large-scale recognition is to delineate complex boundaries of thousands of classes, adequate exploration of feature distributions is important for realizing full potentials of ConvNets. However, state-of-the-art works concentrate only on deeper or wider architecture design, while rarely exploring feature statistics higher than first-order. We take a step towards addressing this problem. Our method consists in covariance pooling, instead of the most commonly used first-order pooling, of highlevel convolutional features. The main challenges involved are robust covariance estimation given a small sample of large-dimensional features and usage of the manifold structure of covariance matrices. To address these challenges, we present a Matrix Power Normalized Covariance (MPNCOV) method. We develop forward and backward propagation formulas regarding the nonlinear matrix functions such that MPN-COV can be trained end-to-end. In addition, we analyze both qualitatively and quantitatively its advantage over the well-known Log-Euclidean metric. On the ImageNet 2012 validation set, by combining MPN-COV we achieve over 4%, 3% and 2.5% gains for AlexNet, VGG-M and VGG-16, respectively; integration of MPN-COV into 50-layer ResNet outperforms ResNet-101 and is comparable to ResNet-152. The source code will be available on the project page: http://www.peihuali.org/MPN-COV." @default.
• W2601789736 created "2017-04-07" @default.
• W2601789736 creator A5018318136 @default.
• W2601789736 creator A5029031269 @default.
• W2601789736 creator A5052272812 @default.
• W2601789736 creator A5070918242 @default.
• W2601789736 date "2017-10-01" @default.
• W2601789736 modified "2023-10-11" @default.
• W2601789736 title "Is Second-Order Information Helpful for Large-Scale Visual Recognition?" @default.
• W2601789736 cites W1677182931 @default.
• W2601789736 cites W1909952827 @default.
• W2601789736 cites W1963882359 @default.
• W2601789736 cites W1966385142 @default.
• W2601789736 cites W1983496390 @default.
• W2601789736 cites W2013912476 @default.
• W2601789736 cites W2029440543 @default.
• W2601789736 cites W2062125287 @default.
• W2601789736 cites W2097117768 @default.
• W2601789736 cites W2104657103 @default.
• W2601789736 cites W2112759033 @default.
• W2601789736 cites W2112796928 @default.
• W2601789736 cites W2117539524 @default.
• W2601789736 cites W2119963163 @default.
• W2601789736 cites W2162915993 @default.
• W2601789736 cites W2194775991 @default.
• W2601789736 cites W2204257188 @default.
• W2601789736 cites W2208314483 @default.
• W2601789736 cites W2324076434 @default.
• W2601789736 cites W2343076010 @default.
• W2601789736 cites W2423822072 @default.
• W2601789736 cites W2582016711 @default.
• W2601789736 doi "https://doi.org/10.1109/iccv.2017.228" @default.
• W2601789736 hasPublicationYear "2017" @default.
• W2601789736 type Work @default.
• W2601789736 sameAs 2601789736 @default.
• W2601789736 citedByCount "186" @default.
• W2601789736 countsByYear W26017897362017 @default.
• W2601789736 countsByYear W26017897362018 @default.
• W2601789736 countsByYear W26017897362019 @default.
• W2601789736 countsByYear W26017897362020 @default.
• W2601789736 countsByYear W26017897362021 @default.
• W2601789736 countsByYear W26017897362022 @default.
• W2601789736 countsByYear W26017897362023 @default.
• W2601789736 crossrefType "proceedings-article" @default.
• W2601789736 hasAuthorship W2601789736A5018318136 @default.
• W2601789736 hasAuthorship W2601789736A5029031269 @default.
• W2601789736 hasAuthorship W2601789736A5052272812 @default.
• W2601789736 hasAuthorship W2601789736A5070918242 @default.
• W2601789736 hasBestOaLocation W26017897362 @default.
• W2601789736 hasConcept C105795698 @default.
• W2601789736 hasConcept C11413529 @default.
• W2601789736 hasConcept C114614502 @default.
• W2601789736 hasConcept C119857082 @default.
• W2601789736 hasConcept C138885662 @default.
• W2601789736 hasConcept C153180895 @default.
• W2601789736 hasConcept C154945302 @default.
• W2601789736 hasConcept C157899210 @default.
• W2601789736 hasConcept C162324750 @default.
• W2601789736 hasConcept C176217482 @default.
• W2601789736 hasConcept C178650346 @default.
• W2601789736 hasConcept C180877172 @default.
• W2601789736 hasConcept C185142706 @default.
• W2601789736 hasConcept C21547014 @default.
• W2601789736 hasConcept C2776401178 @default.
• W2601789736 hasConcept C33923547 @default.
• W2601789736 hasConcept C41008148 @default.
• W2601789736 hasConcept C41895202 @default.
• W2601789736 hasConcept C45347329 @default.
• W2601789736 hasConcept C50644808 @default.
• W2601789736 hasConcept C57273362 @default.
• W2601789736 hasConcept C70437156 @default.
• W2601789736 hasConcept C74193536 @default.
• W2601789736 hasConcept C81363708 @default.
• W2601789736 hasConcept C97931131 @default.
• W2601789736 hasConceptScore W2601789736C105795698 @default.
• W2601789736 hasConceptScore W2601789736C11413529 @default.
• W2601789736 hasConceptScore W2601789736C114614502 @default.
• W2601789736 hasConceptScore W2601789736C119857082 @default.
• W2601789736 hasConceptScore W2601789736C138885662 @default.
• W2601789736 hasConceptScore W2601789736C153180895 @default.
• W2601789736 hasConceptScore W2601789736C154945302 @default.
• W2601789736 hasConceptScore W2601789736C157899210 @default.
• W2601789736 hasConceptScore W2601789736C162324750 @default.
• W2601789736 hasConceptScore W2601789736C176217482 @default.
• W2601789736 hasConceptScore W2601789736C178650346 @default.
• W2601789736 hasConceptScore W2601789736C180877172 @default.
• W2601789736 hasConceptScore W2601789736C185142706 @default.
• W2601789736 hasConceptScore W2601789736C21547014 @default.
• W2601789736 hasConceptScore W2601789736C2776401178 @default.
• W2601789736 hasConceptScore W2601789736C33923547 @default.
• W2601789736 hasConceptScore W2601789736C41008148 @default.
• W2601789736 hasConceptScore W2601789736C41895202 @default.
• W2601789736 hasConceptScore W2601789736C45347329 @default.
• W2601789736 hasConceptScore W2601789736C50644808 @default.
• W2601789736 hasConceptScore W2601789736C57273362 @default.
• W2601789736 hasConceptScore W2601789736C70437156 @default.
• W2601789736 hasConceptScore W2601789736C74193536 @default.
• W2601789736 hasConceptScore W2601789736C81363708 @default.
• W2601789736 hasConceptScore W2601789736C97931131 @default.
• W2601789736 hasLocation W26017897361 @default.

• next