Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385250695> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4385250695 endingPage "6654" @default.
- W4385250695 startingPage "6645" @default.
- W4385250695 abstract "Vision-based Continuous Sign Language Recognition (CSLR) is a challenging and weakly supervised task aimed at segmenting sign language from weakly annotated image stream sequences for recognition. Compared with Isolated Sign Language Recognition (ISLR), the biggest challenge of this work is that the image stream sequences have ambiguous time boundaries. Recent CSLR works have shown that the visual-level sign language recognition task focuses on image stream feature extraction and feature alignment, and overfitting is the most critical problem in the CSLR training process. After investigating the advanced CSLR models in recent years, we have identified that the key to this study is the adequate training of the feature extractor. Therefore, this paper proposes a CSLR model with Multi-state Feature Optimization (MFO), which is based on Fully Convolutional Network (FCN) and Connectionist Temporal Classification (CTC). The MFO mechanism supervises the multiple states of each Sign Gloss in the modeling process and provides more refined labels for training the CTC decoder, which can effectively solve the overfitting problem caused by training, while also significantly reducing the training cost in time. We validate the MFO method on the popular CSLR dataset and demonstrate that the model has better performance." @default.
- W4385250695 created "2023-07-26" @default.
- W4385250695 creator A5007024868 @default.
- W4385250695 creator A5033815877 @default.
- W4385250695 creator A5035923808 @default.
- W4385250695 creator A5050008056 @default.
- W4385250695 creator A5088198922 @default.
- W4385250695 creator A5088930145 @default.
- W4385250695 date "2023-10-04" @default.
- W4385250695 modified "2023-10-09" @default.
- W4385250695 title "Multi-state feature optimization of sign glosses for continuous sign language recognition" @default.
- W4385250695 cites W1995171971 @default.
- W4385250695 cites W2043356651 @default.
- W4385250695 cites W2112382942 @default.
- W4385250695 cites W2127141656 @default.
- W4385250695 cites W2131308039 @default.
- W4385250695 cites W2188882108 @default.
- W4385250695 cites W2587277634 @default.
- W4385250695 cites W2759302818 @default.
- W4385250695 cites W2895638065 @default.
- W4385250695 cites W2896487916 @default.
- W4385250695 cites W2908497602 @default.
- W4385250695 cites W2941870244 @default.
- W4385250695 cites W2948139159 @default.
- W4385250695 cites W2964253156 @default.
- W4385250695 cites W2966344125 @default.
- W4385250695 cites W2979646104 @default.
- W4385250695 cites W3019794902 @default.
- W4385250695 cites W3046952127 @default.
- W4385250695 cites W3092363664 @default.
- W4385250695 cites W3143289910 @default.
- W4385250695 cites W3147467731 @default.
- W4385250695 cites W3160382762 @default.
- W4385250695 cites W3208014149 @default.
- W4385250695 cites W4234870397 @default.
- W4385250695 cites W4311987909 @default.
- W4385250695 cites W950853366 @default.
- W4385250695 doi "https://doi.org/10.3233/jifs-223601" @default.
- W4385250695 hasPublicationYear "2023" @default.
- W4385250695 type Work @default.
- W4385250695 citedByCount "0" @default.
- W4385250695 crossrefType "journal-article" @default.
- W4385250695 hasAuthorship W4385250695A5007024868 @default.
- W4385250695 hasAuthorship W4385250695A5033815877 @default.
- W4385250695 hasAuthorship W4385250695A5035923808 @default.
- W4385250695 hasAuthorship W4385250695A5050008056 @default.
- W4385250695 hasAuthorship W4385250695A5088198922 @default.
- W4385250695 hasAuthorship W4385250695A5088930145 @default.
- W4385250695 hasConcept C138885662 @default.
- W4385250695 hasConcept C153180895 @default.
- W4385250695 hasConcept C154945302 @default.
- W4385250695 hasConcept C22019652 @default.
- W4385250695 hasConcept C2776401178 @default.
- W4385250695 hasConcept C28490314 @default.
- W4385250695 hasConcept C41008148 @default.
- W4385250695 hasConcept C41895202 @default.
- W4385250695 hasConcept C50644808 @default.
- W4385250695 hasConcept C522192633 @default.
- W4385250695 hasConcept C52622490 @default.
- W4385250695 hasConcept C81363708 @default.
- W4385250695 hasConceptScore W4385250695C138885662 @default.
- W4385250695 hasConceptScore W4385250695C153180895 @default.
- W4385250695 hasConceptScore W4385250695C154945302 @default.
- W4385250695 hasConceptScore W4385250695C22019652 @default.
- W4385250695 hasConceptScore W4385250695C2776401178 @default.
- W4385250695 hasConceptScore W4385250695C28490314 @default.
- W4385250695 hasConceptScore W4385250695C41008148 @default.
- W4385250695 hasConceptScore W4385250695C41895202 @default.
- W4385250695 hasConceptScore W4385250695C50644808 @default.
- W4385250695 hasConceptScore W4385250695C522192633 @default.
- W4385250695 hasConceptScore W4385250695C52622490 @default.
- W4385250695 hasConceptScore W4385250695C81363708 @default.
- W4385250695 hasIssue "4" @default.
- W4385250695 hasLocation W43852506951 @default.
- W4385250695 hasOpenAccess W4385250695 @default.
- W4385250695 hasPrimaryLocation W43852506951 @default.
- W4385250695 hasRelatedWork W1492295194 @default.
- W4385250695 hasRelatedWork W1574414179 @default.
- W4385250695 hasRelatedWork W2490526372 @default.
- W4385250695 hasRelatedWork W4221142204 @default.
- W4385250695 hasRelatedWork W4281702477 @default.
- W4385250695 hasRelatedWork W4297676672 @default.
- W4385250695 hasRelatedWork W4362597605 @default.
- W4385250695 hasRelatedWork W4376166922 @default.
- W4385250695 hasRelatedWork W4378510483 @default.
- W4385250695 hasRelatedWork W757031997 @default.
- W4385250695 hasVolume "45" @default.
- W4385250695 isParatext "false" @default.
- W4385250695 isRetracted "false" @default.
- W4385250695 workType "article" @default.