FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2982573303> ?p ?o ?g. }

• W2982573303 abstract "Current image captioning systems perform at a merely descriptive level, essentially enumerating the objects in the scene and their relations. Humans, on the contrary, interpret images by integrating several sources of prior knowledge of the world. In this work, we aim to take a step closer to producing captions that offer a plausible interpretation of the scene, by integrating such contextual information into the captioning pipeline. For this we focus on the captioning of images used to illustrate news articles. We propose a novel captioning method that is able to leverage contextual information provided by the text of news articles associated with an image. Our model is able to selectively draw information from the article guided by visual cues, and to dynamically extend the output dictionary to out-of-vocabulary named entities that appear in the context source. Furthermore we introduce ``GoodNews'', the largest news image captioning dataset in the literature and demonstrate state-of-the-art results." @default.
• W2982573303 created "2019-11-08" @default.
• W2982573303 creator A5008562740 @default.
• W2982573303 creator A5019081944 @default.
• W2982573303 creator A5026710474 @default.
• W2982573303 creator A5029375723 @default.
• W2982573303 date "2019-06-01" @default.
• W2982573303 modified "2023-10-07" @default.
• W2982573303 title "Good News, Everyone! Context Driven Entity-Aware Captioning for News Images" @default.
• W2982573303 cites W1773149199 @default.
• W2982573303 cites W1895577753 @default.
• W2982573303 cites W1931639407 @default.
• W2982573303 cites W1956340063 @default.
• W2982573303 cites W2020278455 @default.
• W2982573303 cites W2105103432 @default.
• W2982573303 cites W2117539524 @default.
• W2982573303 cites W2128856065 @default.
• W2982573303 cites W2133459682 @default.
• W2982573303 cites W2156830856 @default.
• W2982573303 cites W2194775991 @default.
• W2982573303 cites W2250539671 @default.
• W2982573303 cites W2282219577 @default.
• W2982573303 cites W2308864855 @default.
• W2982573303 cites W2325920097 @default.
• W2982573303 cites W2463955103 @default.
• W2982573303 cites W2481240925 @default.
• W2982573303 cites W2506483933 @default.
• W2982573303 cites W2555042572 @default.
• W2982573303 cites W2625940279 @default.
• W2982573303 cites W2735553491 @default.
• W2982573303 cites W2803259101 @default.
• W2982573303 cites W2963084599 @default.
• W2982573303 cites W2963138277 @default.
• W2982573303 cites W2963267809 @default.
• W2982573303 cites W2963305465 @default.
• W2982573303 cites W2964042428 @default.
• W2982573303 doi "https://doi.org/10.1109/cvpr.2019.01275" @default.
• W2982573303 hasPublicationYear "2019" @default.
• W2982573303 type Work @default.
• W2982573303 sameAs 2982573303 @default.
• W2982573303 citedByCount "64" @default.
• W2982573303 countsByYear W29825733032012 @default.
• W2982573303 countsByYear W29825733032019 @default.
• W2982573303 countsByYear W29825733032020 @default.
• W2982573303 countsByYear W29825733032021 @default.
• W2982573303 countsByYear W29825733032022 @default.
• W2982573303 countsByYear W29825733032023 @default.
• W2982573303 crossrefType "proceedings-article" @default.
• W2982573303 hasAuthorship W2982573303A5008562740 @default.
• W2982573303 hasAuthorship W2982573303A5019081944 @default.
• W2982573303 hasAuthorship W2982573303A5026710474 @default.
• W2982573303 hasAuthorship W2982573303A5029375723 @default.
• W2982573303 hasBestOaLocation W29825733032 @default.
• W2982573303 hasConcept C115961682 @default.
• W2982573303 hasConcept C120665830 @default.
• W2982573303 hasConcept C121332964 @default.
• W2982573303 hasConcept C138885662 @default.
• W2982573303 hasConcept C151730666 @default.
• W2982573303 hasConcept C153083717 @default.
• W2982573303 hasConcept C154945302 @default.
• W2982573303 hasConcept C157657479 @default.
• W2982573303 hasConcept C192209626 @default.
• W2982573303 hasConcept C199360897 @default.
• W2982573303 hasConcept C204321447 @default.
• W2982573303 hasConcept C23123220 @default.
• W2982573303 hasConcept C2777601683 @default.
• W2982573303 hasConcept C2779343474 @default.
• W2982573303 hasConcept C36464697 @default.
• W2982573303 hasConcept C41008148 @default.
• W2982573303 hasConcept C41895202 @default.
• W2982573303 hasConcept C43521106 @default.
• W2982573303 hasConcept C86803240 @default.
• W2982573303 hasConceptScore W2982573303C115961682 @default.
• W2982573303 hasConceptScore W2982573303C120665830 @default.
• W2982573303 hasConceptScore W2982573303C121332964 @default.
• W2982573303 hasConceptScore W2982573303C138885662 @default.
• W2982573303 hasConceptScore W2982573303C151730666 @default.
• W2982573303 hasConceptScore W2982573303C153083717 @default.
• W2982573303 hasConceptScore W2982573303C154945302 @default.
• W2982573303 hasConceptScore W2982573303C157657479 @default.
• W2982573303 hasConceptScore W2982573303C192209626 @default.
• W2982573303 hasConceptScore W2982573303C199360897 @default.
• W2982573303 hasConceptScore W2982573303C204321447 @default.
• W2982573303 hasConceptScore W2982573303C23123220 @default.
• W2982573303 hasConceptScore W2982573303C2777601683 @default.
• W2982573303 hasConceptScore W2982573303C2779343474 @default.
• W2982573303 hasConceptScore W2982573303C36464697 @default.
• W2982573303 hasConceptScore W2982573303C41008148 @default.
• W2982573303 hasConceptScore W2982573303C41895202 @default.
• W2982573303 hasConceptScore W2982573303C43521106 @default.
• W2982573303 hasConceptScore W2982573303C86803240 @default.
• W2982573303 hasLocation W29825733031 @default.
• W2982573303 hasLocation W29825733032 @default.
• W2982573303 hasOpenAccess W2982573303 @default.
• W2982573303 hasPrimaryLocation W29825733031 @default.
• W2982573303 hasRelatedWork W2560207749 @default.
• W2982573303 hasRelatedWork W2795359650 @default.
• W2982573303 hasRelatedWork W2923366293 @default.
• W2982573303 hasRelatedWork W2971688299 @default.
• W2982573303 hasRelatedWork W3008515501 @default.

• next