FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897812785> ?p ?o ?g. }

• W2897812785 abstract "Fisher Vector (FV) has been widely used to aggregate the local descriptors of an image into a global representation in large-scale image retrieval. However, FV has limited learning capability and its parameters are mostly fixed after constructing the codebook, which is inflexible and cannot be trained jointly with deep networks. Moreover, the high dimension of FV makes it difficult to be applied in scenarios compact descriptors are needed. In this paper, we propose a novel compact image description scheme based on Fisher network with binary embedding to solve the large-scale image retrieval problem, which consists of two components: a Fisher encoder component and a binary embedding component. Concretely, the Fisher encoder is a trainable neural network functions as the traditional FV, which aggregates the local descriptors into a global representation. And the binary encoder embeds the high-dimensional FV to a binary vector, which outputs the compact global binary descriptor. To learn such a descriptor, we further introduce a novel and effective loss function, in which maximum margin criterion is exploited to minimize the distances of positive pairs, as well as maximizing the distances of negative pairs. Extensive experiments performed on MPEG-7 CDVS benchmarks and ILSVR2010 demonstrate that the proposed framework can achieve very superior performance over the state-of-the-art methods." @default.
• W2897812785 created "2018-10-26" @default.
• W2897812785 creator A5007118591 @default.
• W2897812785 creator A5016080094 @default.
• W2897812785 creator A5057287134 @default.
• W2897812785 creator A5058657142 @default.
• W2897812785 creator A5082338634 @default.
• W2897812785 date "2018-07-01" @default.
• W2897812785 modified "2023-09-23" @default.
• W2897812785 title "Towards Compact Visual Descriptor via Deep Fisher Network with Binary Embedding" @default.
• W2897812785 cites W138476454 @default.
• W2897812785 cites W1606858007 @default.
• W2897812785 cites W1783842908 @default.
• W2897812785 cites W1974647172 @default.
• W2897812785 cites W2005385547 @default.
• W2897812785 cites W2012592962 @default.
• W2897812785 cites W2012833704 @default.
• W2897812785 cites W2071027807 @default.
• W2897812785 cites W2088866137 @default.
• W2897812785 cites W2099027955 @default.
• W2897812785 cites W2108598243 @default.
• W2897812785 cites W2124509324 @default.
• W2897812785 cites W2167244549 @default.
• W2897812785 cites W2170942078 @default.
• W2897812785 cites W2186356322 @default.
• W2897812785 cites W2215369824 @default.
• W2897812785 cites W2292992741 @default.
• W2897812785 cites W2500786414 @default.
• W2897812785 cites W2524324335 @default.
• W2897812785 cites W2744613561 @default.
• W2897812785 cites W280632315 @default.
• W2897812785 cites W2963280127 @default.
• W2897812785 doi "https://doi.org/10.1109/icme.2018.8486520" @default.
• W2897812785 hasPublicationYear "2018" @default.
• W2897812785 type Work @default.
• W2897812785 sameAs 2897812785 @default.
• W2897812785 citedByCount "1" @default.
• W2897812785 countsByYear W28978127852021 @default.
• W2897812785 crossrefType "proceedings-article" @default.
• W2897812785 hasAuthorship W2897812785A5007118591 @default.
• W2897812785 hasAuthorship W2897812785A5016080094 @default.
• W2897812785 hasAuthorship W2897812785A5057287134 @default.
• W2897812785 hasAuthorship W2897812785A5058657142 @default.
• W2897812785 hasAuthorship W2897812785A5082338634 @default.
• W2897812785 hasConcept C111919701 @default.
• W2897812785 hasConcept C11413529 @default.
• W2897812785 hasConcept C115961682 @default.
• W2897812785 hasConcept C118505674 @default.
• W2897812785 hasConcept C119857082 @default.
• W2897812785 hasConcept C127759330 @default.
• W2897812785 hasConcept C153180895 @default.
• W2897812785 hasConcept C154945302 @default.
• W2897812785 hasConcept C1667742 @default.
• W2897812785 hasConcept C17744445 @default.
• W2897812785 hasConcept C199539241 @default.
• W2897812785 hasConcept C2776359362 @default.
• W2897812785 hasConcept C33923547 @default.
• W2897812785 hasConcept C41008148 @default.
• W2897812785 hasConcept C41608201 @default.
• W2897812785 hasConcept C48372109 @default.
• W2897812785 hasConcept C63435697 @default.
• W2897812785 hasConcept C774472 @default.
• W2897812785 hasConcept C94375191 @default.
• W2897812785 hasConcept C94625758 @default.
• W2897812785 hasConceptScore W2897812785C111919701 @default.
• W2897812785 hasConceptScore W2897812785C11413529 @default.
• W2897812785 hasConceptScore W2897812785C115961682 @default.
• W2897812785 hasConceptScore W2897812785C118505674 @default.
• W2897812785 hasConceptScore W2897812785C119857082 @default.
• W2897812785 hasConceptScore W2897812785C127759330 @default.
• W2897812785 hasConceptScore W2897812785C153180895 @default.
• W2897812785 hasConceptScore W2897812785C154945302 @default.
• W2897812785 hasConceptScore W2897812785C1667742 @default.
• W2897812785 hasConceptScore W2897812785C17744445 @default.
• W2897812785 hasConceptScore W2897812785C199539241 @default.
• W2897812785 hasConceptScore W2897812785C2776359362 @default.
• W2897812785 hasConceptScore W2897812785C33923547 @default.
• W2897812785 hasConceptScore W2897812785C41008148 @default.
• W2897812785 hasConceptScore W2897812785C41608201 @default.
• W2897812785 hasConceptScore W2897812785C48372109 @default.
• W2897812785 hasConceptScore W2897812785C63435697 @default.
• W2897812785 hasConceptScore W2897812785C774472 @default.
• W2897812785 hasConceptScore W2897812785C94375191 @default.
• W2897812785 hasConceptScore W2897812785C94625758 @default.
• W2897812785 hasLocation W28978127851 @default.
• W2897812785 hasOpenAccess W2897812785 @default.
• W2897812785 hasPrimaryLocation W28978127851 @default.
• W2897812785 hasRelatedWork W1490587364 @default.
• W2897812785 hasRelatedWork W2030586639 @default.
• W2897812785 hasRelatedWork W2032943964 @default.
• W2897812785 hasRelatedWork W2044521651 @default.
• W2897812785 hasRelatedWork W2204975001 @default.
• W2897812785 hasRelatedWork W2359360576 @default.
• W2897812785 hasRelatedWork W2470592991 @default.
• W2897812785 hasRelatedWork W2484990099 @default.
• W2897812785 hasRelatedWork W2509935066 @default.
• W2897812785 hasRelatedWork W2518953909 @default.
• W2897812785 hasRelatedWork W2528910155 @default.
• W2897812785 hasRelatedWork W2737652487 @default.
• W2897812785 hasRelatedWork W2772926850 @default.

• next