Matches in SemOpenAlex for { <https://semopenalex.org/work/W4304080462> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4304080462 abstract "Stylized Image Captioning aims to generate captions with accurate image content and stylized elements simultaneously. However, large-scaled image and stylized caption pairs cost lots of resources and are usually unavailable. Therefore, it's a challenge to generate stylized captions without paired stylized caption dataset. Previous work on controlling the style of generated captions in an unsupervised way can be divided into two ways: implicitly and explicitly. The former mainly relies on a well-trained language model to capture style knowledge, which is limited to a single style and hard to handle multi-style task. Thus, the latter uses extra style constraints such as outlined style labels or stylized words extracted from stylized sentences to control the style rather than the trained style-specific language model. However, certain styles, such as humorous and romance, are implied in the whole sentence, instead of in some words of a sentence. To address the problems above, we propose a two-step method based on Transformer: firstly detach style representations from large-scaled stylized text-only corpus to provide more holistic style supervision, and secondly attach the style representations to image content to generate stylized captions. We learn a shared image-text space to narrow the gap between the image and the text modality for better attachment. Due to the trade-off between semantics and style, we explore three injection methods of style representations to balance two requirements of image content preservation and stylization. Experiments show that our method outperforms the state-of-the-art systems in overall performance, especially on implied styles." @default.
- W4304080462 created "2022-10-10" @default.
- W4304080462 creator A5001781730 @default.
- W4304080462 creator A5014111141 @default.
- W4304080462 creator A5014223736 @default.
- W4304080462 creator A5015344604 @default.
- W4304080462 creator A5067997634 @default.
- W4304080462 creator A5083893200 @default.
- W4304080462 creator A5087617391 @default.
- W4304080462 date "2022-10-10" @default.
- W4304080462 modified "2023-09-28" @default.
- W4304080462 title "Detach and Attach: Stylized Image Captioning without Paired Stylized Dataset" @default.
- W4304080462 cites W1566289585 @default.
- W4304080462 cites W1631260214 @default.
- W4304080462 cites W1832693441 @default.
- W4304080462 cites W2481240925 @default.
- W4304080462 cites W2822349497 @default.
- W4304080462 cites W2890231609 @default.
- W4304080462 cites W2962917899 @default.
- W4304080462 cites W2969214802 @default.
- W4304080462 cites W2997248215 @default.
- W4304080462 cites W3206055785 @default.
- W4304080462 doi "https://doi.org/10.1145/3503161.3548295" @default.
- W4304080462 hasPublicationYear "2022" @default.
- W4304080462 type Work @default.
- W4304080462 citedByCount "0" @default.
- W4304080462 crossrefType "proceedings-article" @default.
- W4304080462 hasAuthorship W4304080462A5001781730 @default.
- W4304080462 hasAuthorship W4304080462A5014111141 @default.
- W4304080462 hasAuthorship W4304080462A5014223736 @default.
- W4304080462 hasAuthorship W4304080462A5015344604 @default.
- W4304080462 hasAuthorship W4304080462A5067997634 @default.
- W4304080462 hasAuthorship W4304080462A5083893200 @default.
- W4304080462 hasAuthorship W4304080462A5087617391 @default.
- W4304080462 hasBestOaLocation W43040804621 @default.
- W4304080462 hasConcept C115961682 @default.
- W4304080462 hasConcept C124952713 @default.
- W4304080462 hasConcept C139719470 @default.
- W4304080462 hasConcept C142362112 @default.
- W4304080462 hasConcept C154945302 @default.
- W4304080462 hasConcept C157657479 @default.
- W4304080462 hasConcept C162324750 @default.
- W4304080462 hasConcept C204321447 @default.
- W4304080462 hasConcept C2776445246 @default.
- W4304080462 hasConcept C2777530160 @default.
- W4304080462 hasConcept C38935604 @default.
- W4304080462 hasConcept C41008148 @default.
- W4304080462 hasConceptScore W4304080462C115961682 @default.
- W4304080462 hasConceptScore W4304080462C124952713 @default.
- W4304080462 hasConceptScore W4304080462C139719470 @default.
- W4304080462 hasConceptScore W4304080462C142362112 @default.
- W4304080462 hasConceptScore W4304080462C154945302 @default.
- W4304080462 hasConceptScore W4304080462C157657479 @default.
- W4304080462 hasConceptScore W4304080462C162324750 @default.
- W4304080462 hasConceptScore W4304080462C204321447 @default.
- W4304080462 hasConceptScore W4304080462C2776445246 @default.
- W4304080462 hasConceptScore W4304080462C2777530160 @default.
- W4304080462 hasConceptScore W4304080462C38935604 @default.
- W4304080462 hasConceptScore W4304080462C41008148 @default.
- W4304080462 hasFunder F4320321001 @default.
- W4304080462 hasLocation W43040804621 @default.
- W4304080462 hasOpenAccess W4304080462 @default.
- W4304080462 hasPrimaryLocation W43040804621 @default.
- W4304080462 hasRelatedWork W2795359650 @default.
- W4304080462 hasRelatedWork W2822349497 @default.
- W4304080462 hasRelatedWork W2923366293 @default.
- W4304080462 hasRelatedWork W2952640344 @default.
- W4304080462 hasRelatedWork W3008515501 @default.
- W4304080462 hasRelatedWork W3107474891 @default.
- W4304080462 hasRelatedWork W3109516005 @default.
- W4304080462 hasRelatedWork W4226238193 @default.
- W4304080462 hasRelatedWork W4229082270 @default.
- W4304080462 hasRelatedWork W4327773552 @default.
- W4304080462 isParatext "false" @default.
- W4304080462 isRetracted "false" @default.
- W4304080462 workType "article" @default.