Matches in SemOpenAlex for { <https://semopenalex.org/work/W3000202480> ?p ?o ?g. }
- W3000202480 abstract "Visual-semantic embedding enables various tasks such as image-text retrieval, image captioning, and visual question answering. The key to successful visual-semantic embedding is to express visual and textual data properly by accounting for their intricate relationship. While previous studies have achieved much advance by encoding the visual and textual data into a joint space where similar concepts are closely located, they often represent data by a single vector ignoring the presence of multiple important components in an image or text. Thus, in addition to the joint embedding space, we propose a novel multi-head self-attention network to capture various components of visual and textual data by attending to important parts in data. Our approach achieves the new state-of-the-art results in image-text retrieval tasks on MS-COCO and Flicker30K datasets. Through the visualization of the attention maps that capture distinct semantic components at multiple positions in the image and the text, we demonstrate that our method achieves an effective and interpretable visual-semantic joint space." @default.
- W3000202480 created "2020-01-23" @default.
- W3000202480 creator A5003397706 @default.
- W3000202480 creator A5017549954 @default.
- W3000202480 creator A5081523374 @default.
- W3000202480 creator A5081735317 @default.
- W3000202480 date "2020-01-11" @default.
- W3000202480 modified "2023-09-23" @default.
- W3000202480 title "MHSAN: Multi-Head Self-Attention Network for Visual Semantic Embedding" @default.
- W3000202480 cites W1486649854 @default.
- W3000202480 cites W1527575280 @default.
- W3000202480 cites W1536680647 @default.
- W3000202480 cites W1861492603 @default.
- W3000202480 cites W1905882502 @default.
- W3000202480 cites W2064675550 @default.
- W3000202480 cites W2097117768 @default.
- W3000202480 cites W2117539524 @default.
- W3000202480 cites W2185175083 @default.
- W3000202480 cites W2194775991 @default.
- W3000202480 cites W2209647458 @default.
- W3000202480 cites W2415204069 @default.
- W3000202480 cites W2508827254 @default.
- W3000202480 cites W2546696630 @default.
- W3000202480 cites W2597655663 @default.
- W3000202480 cites W2606473278 @default.
- W3000202480 cites W2739181657 @default.
- W3000202480 cites W2770325561 @default.
- W3000202480 cites W2798782720 @default.
- W3000202480 cites W2884585870 @default.
- W3000202480 cites W2938728617 @default.
- W3000202480 cites W2940929270 @default.
- W3000202480 cites W2962964995 @default.
- W3000202480 cites W2963174729 @default.
- W3000202480 cites W2963383024 @default.
- W3000202480 cites W2963403868 @default.
- W3000202480 cites W2963843116 @default.
- W3000202480 cites W2964120214 @default.
- W3000202480 cites W2964276596 @default.
- W3000202480 cites W2964308564 @default.
- W3000202480 doi "https://doi.org/10.48550/arxiv.2001.03712" @default.
- W3000202480 hasPublicationYear "2020" @default.
- W3000202480 type Work @default.
- W3000202480 sameAs 3000202480 @default.
- W3000202480 citedByCount "0" @default.
- W3000202480 crossrefType "posted-content" @default.
- W3000202480 hasAuthorship W3000202480A5003397706 @default.
- W3000202480 hasAuthorship W3000202480A5017549954 @default.
- W3000202480 hasAuthorship W3000202480A5081523374 @default.
- W3000202480 hasAuthorship W3000202480A5081735317 @default.
- W3000202480 hasBestOaLocation W30002024801 @default.
- W3000202480 hasConcept C111919701 @default.
- W3000202480 hasConcept C114793014 @default.
- W3000202480 hasConcept C115961682 @default.
- W3000202480 hasConcept C127313418 @default.
- W3000202480 hasConcept C127413603 @default.
- W3000202480 hasConcept C154945302 @default.
- W3000202480 hasConcept C169760540 @default.
- W3000202480 hasConcept C170154142 @default.
- W3000202480 hasConcept C18555067 @default.
- W3000202480 hasConcept C204321447 @default.
- W3000202480 hasConcept C207363949 @default.
- W3000202480 hasConcept C23123220 @default.
- W3000202480 hasConcept C26517878 @default.
- W3000202480 hasConcept C26760741 @default.
- W3000202480 hasConcept C2778572836 @default.
- W3000202480 hasConcept C2780312720 @default.
- W3000202480 hasConcept C2986420190 @default.
- W3000202480 hasConcept C36464697 @default.
- W3000202480 hasConcept C38652104 @default.
- W3000202480 hasConcept C41008148 @default.
- W3000202480 hasConcept C41608201 @default.
- W3000202480 hasConcept C86803240 @default.
- W3000202480 hasConceptScore W3000202480C111919701 @default.
- W3000202480 hasConceptScore W3000202480C114793014 @default.
- W3000202480 hasConceptScore W3000202480C115961682 @default.
- W3000202480 hasConceptScore W3000202480C127313418 @default.
- W3000202480 hasConceptScore W3000202480C127413603 @default.
- W3000202480 hasConceptScore W3000202480C154945302 @default.
- W3000202480 hasConceptScore W3000202480C169760540 @default.
- W3000202480 hasConceptScore W3000202480C170154142 @default.
- W3000202480 hasConceptScore W3000202480C18555067 @default.
- W3000202480 hasConceptScore W3000202480C204321447 @default.
- W3000202480 hasConceptScore W3000202480C207363949 @default.
- W3000202480 hasConceptScore W3000202480C23123220 @default.
- W3000202480 hasConceptScore W3000202480C26517878 @default.
- W3000202480 hasConceptScore W3000202480C26760741 @default.
- W3000202480 hasConceptScore W3000202480C2778572836 @default.
- W3000202480 hasConceptScore W3000202480C2780312720 @default.
- W3000202480 hasConceptScore W3000202480C2986420190 @default.
- W3000202480 hasConceptScore W3000202480C36464697 @default.
- W3000202480 hasConceptScore W3000202480C38652104 @default.
- W3000202480 hasConceptScore W3000202480C41008148 @default.
- W3000202480 hasConceptScore W3000202480C41608201 @default.
- W3000202480 hasConceptScore W3000202480C86803240 @default.
- W3000202480 hasLocation W30002024801 @default.
- W3000202480 hasOpenAccess W3000202480 @default.
- W3000202480 hasPrimaryLocation W30002024801 @default.
- W3000202480 hasRelatedWork W1578853118 @default.
- W3000202480 hasRelatedWork W2159851394 @default.
- W3000202480 hasRelatedWork W2187131616 @default.