Matches in SemOpenAlex for { <https://semopenalex.org/work/W3018388102> ?p ?o ?g. }
- W3018388102 endingPage "845" @default.
- W3018388102 startingPage "835" @default.
- W3018388102 abstract "Image captioning is a challenging task of visual understanding and has drawn more attention of researchers. In general, two inputs are required at each time step by the Long Short-Term Memory (LSTM) network used in popular attention based image captioning frameworks, including image features and previous generated words. However, error will be accumulated if the previous words are not accurate and the related semantic is not efficient enough. Facing these challenges, a novel model named CaptionNet is proposed in this work as an improved LSTM specially designed for image captioning. Concretely, only attended image features are allowed to be fed into the memory of CaptionNet through input gates. In this way, the dependency on the previous predicted words can be reduced, forcing model to focus on more visual clues of images at the current time step. Moreover, a memory initialization method called image feature encoding is designed to capture richer semantics of the target image. The evaluation on the benchmark MSCOCO and Flickr30K datasets demonstrates the effectiveness of the proposed CaptionNet model, and extensive ablation studies are performed to verify each of the proposed methods. The project page can be found in https://mic.tongji.edu.cn/3f/9c/c9778a147356/page.htm." @default.
- W3018388102 created "2020-05-01" @default.
- W3018388102 creator A5019414690 @default.
- W3018388102 creator A5035895392 @default.
- W3018388102 creator A5058982350 @default.
- W3018388102 creator A5084482894 @default.
- W3018388102 date "2021-01-01" @default.
- W3018388102 modified "2023-10-17" @default.
- W3018388102 title "CaptionNet: A Tailor-made Recurrent Neural Network for Generating Image Descriptions" @default.
- W3018388102 cites W1895577753 @default.
- W3018388102 cites W1905882502 @default.
- W3018388102 cites W1956340063 @default.
- W3018388102 cites W2064675550 @default.
- W3018388102 cites W2097117768 @default.
- W3018388102 cites W2101105183 @default.
- W3018388102 cites W2108325777 @default.
- W3018388102 cites W2112796928 @default.
- W3018388102 cites W2117539524 @default.
- W3018388102 cites W2133459682 @default.
- W3018388102 cites W2157331557 @default.
- W3018388102 cites W2185175083 @default.
- W3018388102 cites W2194775991 @default.
- W3018388102 cites W2220981600 @default.
- W3018388102 cites W2302086703 @default.
- W3018388102 cites W2463955103 @default.
- W3018388102 cites W2506483933 @default.
- W3018388102 cites W2552161745 @default.
- W3018388102 cites W2575842049 @default.
- W3018388102 cites W2579549467 @default.
- W3018388102 cites W2618530766 @default.
- W3018388102 cites W2745461083 @default.
- W3018388102 cites W2754689878 @default.
- W3018388102 cites W2795151422 @default.
- W3018388102 cites W2885013662 @default.
- W3018388102 cites W2887585070 @default.
- W3018388102 cites W2890531016 @default.
- W3018388102 cites W2906314281 @default.
- W3018388102 cites W2913618459 @default.
- W3018388102 cites W2949376505 @default.
- W3018388102 cites W2951183276 @default.
- W3018388102 cites W2963084599 @default.
- W3018388102 cites W2963101956 @default.
- W3018388102 cites W3099884890 @default.
- W3018388102 cites W639708223 @default.
- W3018388102 doi "https://doi.org/10.1109/tmm.2020.2990074" @default.
- W3018388102 hasPublicationYear "2021" @default.
- W3018388102 type Work @default.
- W3018388102 sameAs 3018388102 @default.
- W3018388102 citedByCount "18" @default.
- W3018388102 countsByYear W30183881022021 @default.
- W3018388102 countsByYear W30183881022022 @default.
- W3018388102 countsByYear W30183881022023 @default.
- W3018388102 crossrefType "journal-article" @default.
- W3018388102 hasAuthorship W3018388102A5019414690 @default.
- W3018388102 hasAuthorship W3018388102A5035895392 @default.
- W3018388102 hasAuthorship W3018388102A5058982350 @default.
- W3018388102 hasAuthorship W3018388102A5084482894 @default.
- W3018388102 hasConcept C114466953 @default.
- W3018388102 hasConcept C115961682 @default.
- W3018388102 hasConcept C120665830 @default.
- W3018388102 hasConcept C121332964 @default.
- W3018388102 hasConcept C125411270 @default.
- W3018388102 hasConcept C127313418 @default.
- W3018388102 hasConcept C13280743 @default.
- W3018388102 hasConcept C133488467 @default.
- W3018388102 hasConcept C138885662 @default.
- W3018388102 hasConcept C147168706 @default.
- W3018388102 hasConcept C153180895 @default.
- W3018388102 hasConcept C154945302 @default.
- W3018388102 hasConcept C157657479 @default.
- W3018388102 hasConcept C162324750 @default.
- W3018388102 hasConcept C184337299 @default.
- W3018388102 hasConcept C185798385 @default.
- W3018388102 hasConcept C187736073 @default.
- W3018388102 hasConcept C192209626 @default.
- W3018388102 hasConcept C197115733 @default.
- W3018388102 hasConcept C19768560 @default.
- W3018388102 hasConcept C199360897 @default.
- W3018388102 hasConcept C204321447 @default.
- W3018388102 hasConcept C205649164 @default.
- W3018388102 hasConcept C2776401178 @default.
- W3018388102 hasConcept C2780451532 @default.
- W3018388102 hasConcept C41008148 @default.
- W3018388102 hasConcept C41895202 @default.
- W3018388102 hasConcept C49204034 @default.
- W3018388102 hasConcept C50644808 @default.
- W3018388102 hasConceptScore W3018388102C114466953 @default.
- W3018388102 hasConceptScore W3018388102C115961682 @default.
- W3018388102 hasConceptScore W3018388102C120665830 @default.
- W3018388102 hasConceptScore W3018388102C121332964 @default.
- W3018388102 hasConceptScore W3018388102C125411270 @default.
- W3018388102 hasConceptScore W3018388102C127313418 @default.
- W3018388102 hasConceptScore W3018388102C13280743 @default.
- W3018388102 hasConceptScore W3018388102C133488467 @default.
- W3018388102 hasConceptScore W3018388102C138885662 @default.
- W3018388102 hasConceptScore W3018388102C147168706 @default.
- W3018388102 hasConceptScore W3018388102C153180895 @default.
- W3018388102 hasConceptScore W3018388102C154945302 @default.