Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297102746> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W4297102746 endingPage "108335" @default.
- W4297102746 startingPage "108335" @default.
- W4297102746 abstract "Cross-view geo-localization is to localize the same geographic target in images from different perspectives, e.g., satellite-view and drone-view. The primary challenge faced by existing methods is the large visual appearance changes across views. Most previous work utilizes the deep neural network to obtain the discriminative representations and directly uses them to accomplish the geo-localization task. However, these approaches ignore that the redundancy retained in the extracted features negatively impacts the result. In this paper, we argue that the information bottleneck (IB) can retain the most relevant information while removing as much redundancy as possible. The variational self-distillation (VSD) strategy provides an accurate and analytical solution to estimate the mutual information. To this end, we propose to learn discriminative representations via variational self-distillation (dubbed LDRVSD). Extensive experiments are conducted on two widely-used datasets University-1652 and CVACT, showing the remarkable performance improvements obtained by our LDRVSD method compared with several state-of-the-art approaches." @default.
- W4297102746 created "2022-09-27" @default.
- W4297102746 creator A5009074046 @default.
- W4297102746 creator A5011948808 @default.
- W4297102746 creator A5047946492 @default.
- W4297102746 creator A5064009312 @default.
- W4297102746 creator A5072517057 @default.
- W4297102746 date "2022-10-01" @default.
- W4297102746 modified "2023-09-25" @default.
- W4297102746 title "Learning discriminative representations via variational self-distillation for cross-view geo-localization" @default.
- W4297102746 cites W1938020354 @default.
- W4297102746 cites W1946093182 @default.
- W4297102746 cites W2100799972 @default.
- W4297102746 cites W2108598243 @default.
- W4297102746 cites W2479919622 @default.
- W4297102746 cites W2591669147 @default.
- W4297102746 cites W2620998106 @default.
- W4297102746 cites W2962856082 @default.
- W4297102746 cites W2963474852 @default.
- W4297102746 cites W2997129483 @default.
- W4297102746 cites W3005971801 @default.
- W4297102746 cites W3081227581 @default.
- W4297102746 cites W3115854999 @default.
- W4297102746 cites W3173203577 @default.
- W4297102746 cites W3214744507 @default.
- W4297102746 doi "https://doi.org/10.1016/j.compeleceng.2022.108335" @default.
- W4297102746 hasPublicationYear "2022" @default.
- W4297102746 type Work @default.
- W4297102746 citedByCount "2" @default.
- W4297102746 countsByYear W42971027462023 @default.
- W4297102746 crossrefType "journal-article" @default.
- W4297102746 hasAuthorship W4297102746A5009074046 @default.
- W4297102746 hasAuthorship W4297102746A5011948808 @default.
- W4297102746 hasAuthorship W4297102746A5047946492 @default.
- W4297102746 hasAuthorship W4297102746A5064009312 @default.
- W4297102746 hasAuthorship W4297102746A5072517057 @default.
- W4297102746 hasConcept C111919701 @default.
- W4297102746 hasConcept C119857082 @default.
- W4297102746 hasConcept C149635348 @default.
- W4297102746 hasConcept C152124472 @default.
- W4297102746 hasConcept C152139883 @default.
- W4297102746 hasConcept C153180895 @default.
- W4297102746 hasConcept C154945302 @default.
- W4297102746 hasConcept C178790620 @default.
- W4297102746 hasConcept C185592680 @default.
- W4297102746 hasConcept C204030448 @default.
- W4297102746 hasConcept C2780513914 @default.
- W4297102746 hasConcept C41008148 @default.
- W4297102746 hasConcept C60008888 @default.
- W4297102746 hasConcept C97931131 @default.
- W4297102746 hasConceptScore W4297102746C111919701 @default.
- W4297102746 hasConceptScore W4297102746C119857082 @default.
- W4297102746 hasConceptScore W4297102746C149635348 @default.
- W4297102746 hasConceptScore W4297102746C152124472 @default.
- W4297102746 hasConceptScore W4297102746C152139883 @default.
- W4297102746 hasConceptScore W4297102746C153180895 @default.
- W4297102746 hasConceptScore W4297102746C154945302 @default.
- W4297102746 hasConceptScore W4297102746C178790620 @default.
- W4297102746 hasConceptScore W4297102746C185592680 @default.
- W4297102746 hasConceptScore W4297102746C204030448 @default.
- W4297102746 hasConceptScore W4297102746C2780513914 @default.
- W4297102746 hasConceptScore W4297102746C41008148 @default.
- W4297102746 hasConceptScore W4297102746C60008888 @default.
- W4297102746 hasConceptScore W4297102746C97931131 @default.
- W4297102746 hasLocation W42971027461 @default.
- W4297102746 hasOpenAccess W4297102746 @default.
- W4297102746 hasPrimaryLocation W42971027461 @default.
- W4297102746 hasRelatedWork W1972656095 @default.
- W4297102746 hasRelatedWork W2024160000 @default.
- W4297102746 hasRelatedWork W2061273563 @default.
- W4297102746 hasRelatedWork W2285052147 @default.
- W4297102746 hasRelatedWork W2729514902 @default.
- W4297102746 hasRelatedWork W2743258233 @default.
- W4297102746 hasRelatedWork W2773500201 @default.
- W4297102746 hasRelatedWork W2783047733 @default.
- W4297102746 hasRelatedWork W2998168123 @default.
- W4297102746 hasRelatedWork W4287995534 @default.
- W4297102746 hasVolume "103" @default.
- W4297102746 isParatext "false" @default.
- W4297102746 isRetracted "false" @default.
- W4297102746 workType "article" @default.