Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385488757> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4385488757 abstract "Image captioning has long been widely regarded as a modal transformation task from visual to linguistic modality. Most current research focuses on the information transformation between single modalities dominated by visual features, while less attention is paid to the interaction between visual features and linguistic features. This rigid single-modal conversion method is prone to information confusion and loss during the conversion process, making it difficult for the model to generate accurate and detailed captions. In this paper, we propose a general Semantic-Augmented Transformer-Based (SAT) framework to facilitate smoother transformation between the two modalities. In the encoding stage, we use the fine-grained description of each region to fuse with the corresponding image features to make the image feature representation closer to the text feature representation. In the decoding stage, the caption's part-of-speech information is used as prior knowledge to constrain the model to pay more attention to the details in the image rather than only to the prominent entities for fine-grained captions. We extensively evaluate our framework on various state-of-the-art transformer-based models. Experiments demonstrate that these models have superior performance on the MS-COCO dataset under our framework." @default.
- W4385488757 created "2023-08-03" @default.
- W4385488757 creator A5002262984 @default.
- W4385488757 creator A5049949859 @default.
- W4385488757 creator A5071451609 @default.
- W4385488757 date "2023-06-18" @default.
- W4385488757 modified "2023-09-26" @default.
- W4385488757 title "Image Alone Are Not Enough: A General Semantic-Augmented Transformer-Based Framework for Image Captioning" @default.
- W4385488757 cites W1895577753 @default.
- W4385488757 cites W1905882502 @default.
- W4385488757 cites W1956340063 @default.
- W4385488757 cites W2101105183 @default.
- W4385488757 cites W2558834163 @default.
- W4385488757 cites W2560645892 @default.
- W4385488757 cites W2575842049 @default.
- W4385488757 cites W2745461083 @default.
- W4385488757 cites W2754927243 @default.
- W4385488757 cites W2890531016 @default.
- W4385488757 cites W2896348597 @default.
- W4385488757 cites W2901988662 @default.
- W4385488757 cites W2963101956 @default.
- W4385488757 cites W2965697393 @default.
- W4385488757 cites W2972897806 @default.
- W4385488757 cites W2986670728 @default.
- W4385488757 cites W2990818246 @default.
- W4385488757 cites W2992478697 @default.
- W4385488757 cites W3034655362 @default.
- W4385488757 cites W3034984754 @default.
- W4385488757 cites W3035284526 @default.
- W4385488757 cites W3035497460 @default.
- W4385488757 cites W3167939936 @default.
- W4385488757 cites W3174377922 @default.
- W4385488757 cites W3205071568 @default.
- W4385488757 doi "https://doi.org/10.1109/ijcnn54540.2023.10191656" @default.
- W4385488757 hasPublicationYear "2023" @default.
- W4385488757 type Work @default.
- W4385488757 citedByCount "0" @default.
- W4385488757 crossrefType "proceedings-article" @default.
- W4385488757 hasAuthorship W4385488757A5002262984 @default.
- W4385488757 hasAuthorship W4385488757A5049949859 @default.
- W4385488757 hasAuthorship W4385488757A5071451609 @default.
- W4385488757 hasConcept C115961682 @default.
- W4385488757 hasConcept C121332964 @default.
- W4385488757 hasConcept C125411270 @default.
- W4385488757 hasConcept C138885662 @default.
- W4385488757 hasConcept C144024400 @default.
- W4385488757 hasConcept C154945302 @default.
- W4385488757 hasConcept C157657479 @default.
- W4385488757 hasConcept C165801399 @default.
- W4385488757 hasConcept C204321447 @default.
- W4385488757 hasConcept C2776401178 @default.
- W4385488757 hasConcept C2779903281 @default.
- W4385488757 hasConcept C36289849 @default.
- W4385488757 hasConcept C36464697 @default.
- W4385488757 hasConcept C41008148 @default.
- W4385488757 hasConcept C41895202 @default.
- W4385488757 hasConcept C57273362 @default.
- W4385488757 hasConcept C62520636 @default.
- W4385488757 hasConcept C66322947 @default.
- W4385488757 hasConcept C76155785 @default.
- W4385488757 hasConceptScore W4385488757C115961682 @default.
- W4385488757 hasConceptScore W4385488757C121332964 @default.
- W4385488757 hasConceptScore W4385488757C125411270 @default.
- W4385488757 hasConceptScore W4385488757C138885662 @default.
- W4385488757 hasConceptScore W4385488757C144024400 @default.
- W4385488757 hasConceptScore W4385488757C154945302 @default.
- W4385488757 hasConceptScore W4385488757C157657479 @default.
- W4385488757 hasConceptScore W4385488757C165801399 @default.
- W4385488757 hasConceptScore W4385488757C204321447 @default.
- W4385488757 hasConceptScore W4385488757C2776401178 @default.
- W4385488757 hasConceptScore W4385488757C2779903281 @default.
- W4385488757 hasConceptScore W4385488757C36289849 @default.
- W4385488757 hasConceptScore W4385488757C36464697 @default.
- W4385488757 hasConceptScore W4385488757C41008148 @default.
- W4385488757 hasConceptScore W4385488757C41895202 @default.
- W4385488757 hasConceptScore W4385488757C57273362 @default.
- W4385488757 hasConceptScore W4385488757C62520636 @default.
- W4385488757 hasConceptScore W4385488757C66322947 @default.
- W4385488757 hasConceptScore W4385488757C76155785 @default.
- W4385488757 hasLocation W43854887571 @default.
- W4385488757 hasOpenAccess W4385488757 @default.
- W4385488757 hasPrimaryLocation W43854887571 @default.
- W4385488757 hasRelatedWork W2464179670 @default.
- W4385488757 hasRelatedWork W2547835662 @default.
- W4385488757 hasRelatedWork W2901467237 @default.
- W4385488757 hasRelatedWork W3183824823 @default.
- W4385488757 hasRelatedWork W4200208799 @default.
- W4385488757 hasRelatedWork W4307856881 @default.
- W4385488757 hasRelatedWork W4320016117 @default.
- W4385488757 hasRelatedWork W4320858200 @default.
- W4385488757 hasRelatedWork W4365441642 @default.
- W4385488757 hasRelatedWork W4378907066 @default.
- W4385488757 isParatext "false" @default.
- W4385488757 isRetracted "false" @default.
- W4385488757 workType "article" @default.