Matches in SemOpenAlex for { <https://semopenalex.org/work/W4206210197> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W4206210197 endingPage "15266" @default.
- W4206210197 startingPage "15256" @default.
- W4206210197 abstract "Lightweight models are pivotal in efficient semantic segmentation, but they often suffer from insufficient context information due to limited convolution and small receptive field. To address this problem, we propose a tailored approach to efficient semantic segmentation by leveraging two complementary distillation schemes for supplementing context information to small networks: 1) a self-attention distillation scheme, which transfers long-range context knowledge adaptively from large teacher networks to small student networks; and 2) a layer-wise context distillation scheme, which transfers structured context from deep layers to shallow layers within student networks for promoting semantic consistency of the shallow layers. Extensive experiments on the ADE20K, Cityscapes, and Camvid datasets well demonstrate the effectiveness of our proposal." @default.
- W4206210197 created "2022-01-25" @default.
- W4206210197 creator A5009239895 @default.
- W4206210197 creator A5046787619 @default.
- W4206210197 creator A5079361172 @default.
- W4206210197 creator A5089642905 @default.
- W4206210197 date "2022-09-01" @default.
- W4206210197 modified "2023-10-14" @default.
- W4206210197 title "Efficient Semantic Segmentation via Self-Attention and Self-Distillation" @default.
- W4206210197 cites W1913356549 @default.
- W4206210197 cites W2019329118 @default.
- W4206210197 cites W2124592697 @default.
- W4206210197 cites W2194775991 @default.
- W4206210197 cites W2204696980 @default.
- W4206210197 cites W2340897893 @default.
- W4206210197 cites W2395611524 @default.
- W4206210197 cites W2412782625 @default.
- W4206210197 cites W2560023338 @default.
- W4206210197 cites W2737258237 @default.
- W4206210197 cites W2750432752 @default.
- W4206210197 cites W2778955544 @default.
- W4206210197 cites W2799213142 @default.
- W4206210197 cites W2952787292 @default.
- W4206210197 cites W2954054736 @default.
- W4206210197 cites W2955058313 @default.
- W4206210197 cites W2963091558 @default.
- W4206210197 cites W2963163009 @default.
- W4206210197 cites W2963166928 @default.
- W4206210197 cites W2963418739 @default.
- W4206210197 cites W2963727650 @default.
- W4206210197 cites W2963881378 @default.
- W4206210197 cites W2971198903 @default.
- W4206210197 cites W2981441441 @default.
- W4206210197 cites W2981689412 @default.
- W4206210197 cites W2987861506 @default.
- W4206210197 cites W2997006708 @default.
- W4206210197 cites W3049435445 @default.
- W4206210197 cites W3186802005 @default.
- W4206210197 doi "https://doi.org/10.1109/tits.2021.3139001" @default.
- W4206210197 hasPublicationYear "2022" @default.
- W4206210197 type Work @default.
- W4206210197 citedByCount "12" @default.
- W4206210197 countsByYear W42062101972022 @default.
- W4206210197 countsByYear W42062101972023 @default.
- W4206210197 crossrefType "journal-article" @default.
- W4206210197 hasAuthorship W4206210197A5009239895 @default.
- W4206210197 hasAuthorship W4206210197A5046787619 @default.
- W4206210197 hasAuthorship W4206210197A5079361172 @default.
- W4206210197 hasAuthorship W4206210197A5089642905 @default.
- W4206210197 hasBestOaLocation W42062101972 @default.
- W4206210197 hasConcept C119857082 @default.
- W4206210197 hasConcept C127313418 @default.
- W4206210197 hasConcept C151730666 @default.
- W4206210197 hasConcept C154945302 @default.
- W4206210197 hasConcept C178790620 @default.
- W4206210197 hasConcept C185592680 @default.
- W4206210197 hasConcept C204030448 @default.
- W4206210197 hasConcept C2776436953 @default.
- W4206210197 hasConcept C2779343474 @default.
- W4206210197 hasConcept C41008148 @default.
- W4206210197 hasConcept C45347329 @default.
- W4206210197 hasConcept C50644808 @default.
- W4206210197 hasConcept C89600930 @default.
- W4206210197 hasConceptScore W4206210197C119857082 @default.
- W4206210197 hasConceptScore W4206210197C127313418 @default.
- W4206210197 hasConceptScore W4206210197C151730666 @default.
- W4206210197 hasConceptScore W4206210197C154945302 @default.
- W4206210197 hasConceptScore W4206210197C178790620 @default.
- W4206210197 hasConceptScore W4206210197C185592680 @default.
- W4206210197 hasConceptScore W4206210197C204030448 @default.
- W4206210197 hasConceptScore W4206210197C2776436953 @default.
- W4206210197 hasConceptScore W4206210197C2779343474 @default.
- W4206210197 hasConceptScore W4206210197C41008148 @default.
- W4206210197 hasConceptScore W4206210197C45347329 @default.
- W4206210197 hasConceptScore W4206210197C50644808 @default.
- W4206210197 hasConceptScore W4206210197C89600930 @default.
- W4206210197 hasFunder F4320321001 @default.
- W4206210197 hasFunder F4320335972 @default.
- W4206210197 hasIssue "9" @default.
- W4206210197 hasLocation W42062101971 @default.
- W4206210197 hasLocation W42062101972 @default.
- W4206210197 hasLocation W42062101973 @default.
- W4206210197 hasOpenAccess W4206210197 @default.
- W4206210197 hasPrimaryLocation W42062101971 @default.
- W4206210197 hasRelatedWork W1603736412 @default.
- W4206210197 hasRelatedWork W19111321 @default.
- W4206210197 hasRelatedWork W2344382886 @default.
- W4206210197 hasRelatedWork W2412887479 @default.
- W4206210197 hasRelatedWork W2530481740 @default.
- W4206210197 hasRelatedWork W2768175398 @default.
- W4206210197 hasRelatedWork W2890372105 @default.
- W4206210197 hasRelatedWork W3026162553 @default.
- W4206210197 hasRelatedWork W4304185162 @default.
- W4206210197 hasRelatedWork W4379231730 @default.
- W4206210197 hasVolume "23" @default.
- W4206210197 isParatext "false" @default.
- W4206210197 isRetracted "false" @default.
- W4206210197 workType "article" @default.