Matches in SemOpenAlex for { <https://semopenalex.org/work/W2623164477> ?p ?o ?g. }
- W2623164477 abstract "Numerous embedding models have been recently explored to incorporate semantic knowledge into visual recognition. Existing methods typically focus on minimizing the distance between the corresponding images and texts in the embedding space but do not explicitly optimize the underlying structure. Our key observation is that modeling the pairwise image-image relationship improves the discrimination ability of the embedding model. In this paper, we propose the structured discriminative and difference constraints to learn visual-semantic embeddings. First, we exploit the discriminative constraints to capture the intra- and inter-class relationships of image embeddings. The discriminative constraints encourage separability for image instances of different classes. Second, we align the difference vector between a pair of image embeddings with that of the corresponding word embeddings. The difference constraints help regularize image embeddings to preserve the semantic relationships among word embeddings. Extensive evaluations demonstrate the effectiveness of the proposed structured embeddings for single-label classification, multi-label classification, and zero-shot recognition." @default.
- W2623164477 created "2017-06-15" @default.
- W2623164477 creator A5030965866 @default.
- W2623164477 creator A5051688016 @default.
- W2623164477 creator A5053730462 @default.
- W2623164477 creator A5063179713 @default.
- W2623164477 creator A5088961773 @default.
- W2623164477 date "2017-06-05" @default.
- W2623164477 modified "2023-09-25" @default.
- W2623164477 title "Learning Structured Semantic Embeddings for Visual Recognition." @default.
- W2623164477 cites W1499991161 @default.
- W2623164477 cites W1514027499 @default.
- W2623164477 cites W1686810756 @default.
- W2623164477 cites W1905882502 @default.
- W2623164477 cites W1960364170 @default.
- W2623164477 cites W1988719066 @default.
- W2623164477 cites W2031489346 @default.
- W2623164477 cites W2077071968 @default.
- W2623164477 cites W2090575037 @default.
- W2623164477 cites W2097117768 @default.
- W2623164477 cites W2098411764 @default.
- W2623164477 cites W2108598243 @default.
- W2623164477 cites W2112912048 @default.
- W2623164477 cites W2123024445 @default.
- W2623164477 cites W2146937173 @default.
- W2623164477 cites W2153579005 @default.
- W2623164477 cites W2155893237 @default.
- W2623164477 cites W2171361956 @default.
- W2623164477 cites W219040644 @default.
- W2623164477 cites W2400717490 @default.
- W2623164477 cites W2405223529 @default.
- W2623164477 cites W2441255125 @default.
- W2623164477 cites W2518962550 @default.
- W2623164477 cites W2519967618 @default.
- W2623164477 cites W2527238922 @default.
- W2623164477 cites W2561940122 @default.
- W2623164477 cites W261873710 @default.
- W2623164477 cites W2950276680 @default.
- W2623164477 cites W2962830213 @default.
- W2623164477 cites W2963177757 @default.
- W2623164477 cites W2963389687 @default.
- W2623164477 cites W2963676207 @default.
- W2623164477 cites W2963871344 @default.
- W2623164477 cites W2963899908 @default.
- W2623164477 cites W2964189431 @default.
- W2623164477 cites W2964350399 @default.
- W2623164477 cites W3118608800 @default.
- W2623164477 cites W652269744 @default.
- W2623164477 cites W93016980 @default.
- W2623164477 hasPublicationYear "2017" @default.
- W2623164477 type Work @default.
- W2623164477 sameAs 2623164477 @default.
- W2623164477 citedByCount "4" @default.
- W2623164477 countsByYear W26231644772019 @default.
- W2623164477 countsByYear W26231644772020 @default.
- W2623164477 countsByYear W26231644772021 @default.
- W2623164477 crossrefType "posted-content" @default.
- W2623164477 hasAuthorship W2623164477A5030965866 @default.
- W2623164477 hasAuthorship W2623164477A5051688016 @default.
- W2623164477 hasAuthorship W2623164477A5053730462 @default.
- W2623164477 hasAuthorship W2623164477A5063179713 @default.
- W2623164477 hasAuthorship W2623164477A5088961773 @default.
- W2623164477 hasConcept C115961682 @default.
- W2623164477 hasConcept C119857082 @default.
- W2623164477 hasConcept C120665830 @default.
- W2623164477 hasConcept C121332964 @default.
- W2623164477 hasConcept C153180895 @default.
- W2623164477 hasConcept C154945302 @default.
- W2623164477 hasConcept C165696696 @default.
- W2623164477 hasConcept C1667742 @default.
- W2623164477 hasConcept C184898388 @default.
- W2623164477 hasConcept C192209626 @default.
- W2623164477 hasConcept C204321447 @default.
- W2623164477 hasConcept C2524010 @default.
- W2623164477 hasConcept C26517878 @default.
- W2623164477 hasConcept C2777212361 @default.
- W2623164477 hasConcept C33923547 @default.
- W2623164477 hasConcept C38652104 @default.
- W2623164477 hasConcept C41008148 @default.
- W2623164477 hasConcept C41608201 @default.
- W2623164477 hasConcept C75294576 @default.
- W2623164477 hasConcept C86034646 @default.
- W2623164477 hasConcept C90805587 @default.
- W2623164477 hasConcept C97931131 @default.
- W2623164477 hasConceptScore W2623164477C115961682 @default.
- W2623164477 hasConceptScore W2623164477C119857082 @default.
- W2623164477 hasConceptScore W2623164477C120665830 @default.
- W2623164477 hasConceptScore W2623164477C121332964 @default.
- W2623164477 hasConceptScore W2623164477C153180895 @default.
- W2623164477 hasConceptScore W2623164477C154945302 @default.
- W2623164477 hasConceptScore W2623164477C165696696 @default.
- W2623164477 hasConceptScore W2623164477C1667742 @default.
- W2623164477 hasConceptScore W2623164477C184898388 @default.
- W2623164477 hasConceptScore W2623164477C192209626 @default.
- W2623164477 hasConceptScore W2623164477C204321447 @default.
- W2623164477 hasConceptScore W2623164477C2524010 @default.
- W2623164477 hasConceptScore W2623164477C26517878 @default.
- W2623164477 hasConceptScore W2623164477C2777212361 @default.
- W2623164477 hasConceptScore W2623164477C33923547 @default.
- W2623164477 hasConceptScore W2623164477C38652104 @default.