Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386075496> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4386075496 abstract "Sign languages are visual languages which convey in-formation by signers' handshape, facial expression, body movement, and so forth. Due to the inherent restriction of combinations of these visual ingredients, there exist a significant number of visually indistinguishable signs (VISigns) in sign languages, which limits the recognition capacity of vision neural networks. To mitigate the problem, we propose the Natural Language-Assisted Sign Language Recognition (NLA-SLR) framework, which exploits semantic information contained in glosses (sign labels). First, for VISigns with similar semantic meanings, we propose language-aware label smoothing by generating soft labels for each training sign whose smoothing weights are computed from the normalized semantic similarities among the glosses to ease training. Second, for VISigns with distinct semantic meanings, we present an inter-modality mixup technique which blends vision and gloss features to further maximize the separability of different signs under the super-vision of blended labels. Besides, we also introduce a novel backbone, video-keypoint network, which not only models both RGB videos and human body keypoints but also derives knowledge from sign videos of different temporal receptive fields. Empirically, our method achieves state-of-the-art performance on three widely-adopted benchmarks: MSASL, WLASL, and NMFs-CSL. Codes are available at https://github.com/FangyunWeilSLRT." @default.
- W4386075496 created "2023-08-23" @default.
- W4386075496 creator A5034553547 @default.
- W4386075496 creator A5059141717 @default.
- W4386075496 creator A5090973869 @default.
- W4386075496 date "2023-06-01" @default.
- W4386075496 modified "2023-10-01" @default.
- W4386075496 title "Natural Language-Assisted Sign Language Recognition" @default.
- W4386075496 cites W2064675550 @default.
- W4386075496 cites W2250539671 @default.
- W4386075496 cites W2493916176 @default.
- W4386075496 cites W2963076818 @default.
- W4386075496 cites W2963820951 @default.
- W4386075496 cites W2963855133 @default.
- W4386075496 cites W2990152177 @default.
- W4386075496 cites W2990503944 @default.
- W4386075496 cites W2997931247 @default.
- W4386075496 cites W3009828227 @default.
- W4386075496 cites W3034269985 @default.
- W4386075496 cites W3108425892 @default.
- W4386075496 cites W3135189994 @default.
- W4386075496 cites W3147467731 @default.
- W4386075496 cites W3173262825 @default.
- W4386075496 cites W3184215204 @default.
- W4386075496 cites W3185538031 @default.
- W4386075496 cites W3198377975 @default.
- W4386075496 cites W3202747033 @default.
- W4386075496 cites W3205234797 @default.
- W4386075496 cites W3206600927 @default.
- W4386075496 cites W3216156094 @default.
- W4386075496 cites W4214541243 @default.
- W4386075496 cites W4214673031 @default.
- W4386075496 cites W4214926101 @default.
- W4386075496 cites W4226058394 @default.
- W4386075496 cites W4296068613 @default.
- W4386075496 cites W4312245820 @default.
- W4386075496 cites W4312333224 @default.
- W4386075496 cites W4312648273 @default.
- W4386075496 cites W4312878209 @default.
- W4386075496 cites W4312910375 @default.
- W4386075496 cites W4312916580 @default.
- W4386075496 cites W4312938052 @default.
- W4386075496 cites W4312980231 @default.
- W4386075496 cites W4382240029 @default.
- W4386075496 doi "https://doi.org/10.1109/cvpr52729.2023.01430" @default.
- W4386075496 hasPublicationYear "2023" @default.
- W4386075496 type Work @default.
- W4386075496 citedByCount "1" @default.
- W4386075496 countsByYear W43860754962023 @default.
- W4386075496 crossrefType "proceedings-article" @default.
- W4386075496 hasAuthorship W4386075496A5034553547 @default.
- W4386075496 hasAuthorship W4386075496A5059141717 @default.
- W4386075496 hasAuthorship W4386075496A5090973869 @default.
- W4386075496 hasConcept C138885662 @default.
- W4386075496 hasConcept C154945302 @default.
- W4386075496 hasConcept C195324797 @default.
- W4386075496 hasConcept C204321447 @default.
- W4386075496 hasConcept C2780226545 @default.
- W4386075496 hasConcept C28490314 @default.
- W4386075496 hasConcept C31972630 @default.
- W4386075496 hasConcept C3770464 @default.
- W4386075496 hasConcept C41008148 @default.
- W4386075496 hasConcept C41895202 @default.
- W4386075496 hasConcept C522192633 @default.
- W4386075496 hasConceptScore W4386075496C138885662 @default.
- W4386075496 hasConceptScore W4386075496C154945302 @default.
- W4386075496 hasConceptScore W4386075496C195324797 @default.
- W4386075496 hasConceptScore W4386075496C204321447 @default.
- W4386075496 hasConceptScore W4386075496C2780226545 @default.
- W4386075496 hasConceptScore W4386075496C28490314 @default.
- W4386075496 hasConceptScore W4386075496C31972630 @default.
- W4386075496 hasConceptScore W4386075496C3770464 @default.
- W4386075496 hasConceptScore W4386075496C41008148 @default.
- W4386075496 hasConceptScore W4386075496C41895202 @default.
- W4386075496 hasConceptScore W4386075496C522192633 @default.
- W4386075496 hasLocation W43860754961 @default.
- W4386075496 hasOpenAccess W4386075496 @default.
- W4386075496 hasPrimaryLocation W43860754961 @default.
- W4386075496 hasRelatedWork W159132833 @default.
- W4386075496 hasRelatedWork W2293457016 @default.
- W4386075496 hasRelatedWork W2408420624 @default.
- W4386075496 hasRelatedWork W2502722637 @default.
- W4386075496 hasRelatedWork W2789919619 @default.
- W4386075496 hasRelatedWork W2977842567 @default.
- W4386075496 hasRelatedWork W3198474835 @default.
- W4386075496 hasRelatedWork W87581401 @default.
- W4386075496 hasRelatedWork W1551406738 @default.
- W4386075496 hasRelatedWork W1872130062 @default.
- W4386075496 isParatext "false" @default.
- W4386075496 isRetracted "false" @default.
- W4386075496 workType "article" @default.