Matches in SemOpenAlex for { <https://semopenalex.org/work/W4280496911> ?p ?o ?g. }
- W4280496911 endingPage "102238" @default.
- W4280496911 startingPage "102238" @default.
- W4280496911 abstract "Machine attention mechanisms are widely used in the task of image captioning. Such mechanisms dynamically focus on different regions to guide the word generation process. However, existing attention models may fail to concentrate on correct regions and mislead the word prediction without explicit supervision. In this study, we exploit the human captioning attention encoding rich information that human beings perceive during captioning, and propose a novel Hybrid Attention Network (HAN) that incorporates the prevailing machine attention mechanisms with human captioning attention. The proposed HAN addresses the problem of “object hallucination” by re-weighting bottom-up attention, and improves the diversity of the generated captioning by complementing top-down attention with human captioning attention. Extensive experiments are conducted on Flickr30K and MS COCO datasets, demonstrating that the proposed method effectively improves the performance of the current image captioning methods." @default.
- W4280496911 created "2022-05-22" @default.
- W4280496911 creator A5004651580 @default.
- W4280496911 creator A5030344445 @default.
- W4280496911 creator A5063013411 @default.
- W4280496911 creator A5075080019 @default.
- W4280496911 creator A5085233086 @default.
- W4280496911 date "2022-07-01" @default.
- W4280496911 modified "2023-10-14" @default.
- W4280496911 title "Hybrid attention network for image captioning" @default.
- W4280496911 cites W1892884262 @default.
- W4280496911 cites W1905882502 @default.
- W4280496911 cites W1956340063 @default.
- W4280496911 cites W2100379672 @default.
- W4280496911 cites W2185175083 @default.
- W4280496911 cites W2533370895 @default.
- W4280496911 cites W2594648402 @default.
- W4280496911 cites W2745461083 @default.
- W4280496911 cites W2761635216 @default.
- W4280496911 cites W2962982762 @default.
- W4280496911 cites W2963084599 @default.
- W4280496911 cites W2968101724 @default.
- W4280496911 cites W2987327987 @default.
- W4280496911 cites W2997316506 @default.
- W4280496911 cites W3000351820 @default.
- W4280496911 cites W3019533076 @default.
- W4280496911 cites W3022710784 @default.
- W4280496911 cites W3035323998 @default.
- W4280496911 cites W3119689583 @default.
- W4280496911 cites W3173220247 @default.
- W4280496911 cites W3177244742 @default.
- W4280496911 cites W3188318562 @default.
- W4280496911 cites W3193755308 @default.
- W4280496911 cites W3195588784 @default.
- W4280496911 cites W3200654829 @default.
- W4280496911 cites W3204380095 @default.
- W4280496911 cites W3206086363 @default.
- W4280496911 cites W4205884078 @default.
- W4280496911 doi "https://doi.org/10.1016/j.displa.2022.102238" @default.
- W4280496911 hasPublicationYear "2022" @default.
- W4280496911 type Work @default.
- W4280496911 citedByCount "6" @default.
- W4280496911 countsByYear W42804969112022 @default.
- W4280496911 countsByYear W42804969112023 @default.
- W4280496911 crossrefType "journal-article" @default.
- W4280496911 hasAuthorship W4280496911A5004651580 @default.
- W4280496911 hasAuthorship W4280496911A5030344445 @default.
- W4280496911 hasAuthorship W4280496911A5063013411 @default.
- W4280496911 hasAuthorship W4280496911A5075080019 @default.
- W4280496911 hasAuthorship W4280496911A5085233086 @default.
- W4280496911 hasConcept C111919701 @default.
- W4280496911 hasConcept C115961682 @default.
- W4280496911 hasConcept C120665830 @default.
- W4280496911 hasConcept C121332964 @default.
- W4280496911 hasConcept C126838900 @default.
- W4280496911 hasConcept C127413603 @default.
- W4280496911 hasConcept C138885662 @default.
- W4280496911 hasConcept C154945302 @default.
- W4280496911 hasConcept C157657479 @default.
- W4280496911 hasConcept C183115368 @default.
- W4280496911 hasConcept C192209626 @default.
- W4280496911 hasConcept C201995342 @default.
- W4280496911 hasConcept C204321447 @default.
- W4280496911 hasConcept C2780451532 @default.
- W4280496911 hasConcept C2781238097 @default.
- W4280496911 hasConcept C28490314 @default.
- W4280496911 hasConcept C41008148 @default.
- W4280496911 hasConcept C41895202 @default.
- W4280496911 hasConcept C71924100 @default.
- W4280496911 hasConcept C90805587 @default.
- W4280496911 hasConcept C98045186 @default.
- W4280496911 hasConceptScore W4280496911C111919701 @default.
- W4280496911 hasConceptScore W4280496911C115961682 @default.
- W4280496911 hasConceptScore W4280496911C120665830 @default.
- W4280496911 hasConceptScore W4280496911C121332964 @default.
- W4280496911 hasConceptScore W4280496911C126838900 @default.
- W4280496911 hasConceptScore W4280496911C127413603 @default.
- W4280496911 hasConceptScore W4280496911C138885662 @default.
- W4280496911 hasConceptScore W4280496911C154945302 @default.
- W4280496911 hasConceptScore W4280496911C157657479 @default.
- W4280496911 hasConceptScore W4280496911C183115368 @default.
- W4280496911 hasConceptScore W4280496911C192209626 @default.
- W4280496911 hasConceptScore W4280496911C201995342 @default.
- W4280496911 hasConceptScore W4280496911C204321447 @default.
- W4280496911 hasConceptScore W4280496911C2780451532 @default.
- W4280496911 hasConceptScore W4280496911C2781238097 @default.
- W4280496911 hasConceptScore W4280496911C28490314 @default.
- W4280496911 hasConceptScore W4280496911C41008148 @default.
- W4280496911 hasConceptScore W4280496911C41895202 @default.
- W4280496911 hasConceptScore W4280496911C71924100 @default.
- W4280496911 hasConceptScore W4280496911C90805587 @default.
- W4280496911 hasConceptScore W4280496911C98045186 @default.
- W4280496911 hasFunder F4320321001 @default.
- W4280496911 hasFunder F4320327780 @default.
- W4280496911 hasLocation W42804969111 @default.
- W4280496911 hasOpenAccess W4280496911 @default.
- W4280496911 hasPrimaryLocation W42804969111 @default.
- W4280496911 hasRelatedWork W2945836425 @default.