Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225270920> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4225270920 abstract "This paper proposes a method of detecting dementia from the simultaneous speech and eye-tracking recordings of subjects in a picture description task. First, automatic speech recognition (ASR) and regional picture recognition (RPR) models are built to extract content-related bottleneck (BN) features for both speech and eye-tracking inputs. Then, a neural network is designed to fuse these two modals for discriminating dementia patients from healthy controls. The network contains a cross-modal Transformer encoder for bimodal interaction, and a self-attention Transformer encoder for final classification. Experimental results demonstrate that the detection accuracy of the proposed method is 84.26%, which outperforms baseline methods and ablated models using single speech or eye-tracking input." @default.
- W4225270920 created "2022-05-04" @default.
- W4225270920 creator A5004434747 @default.
- W4225270920 creator A5007911614 @default.
- W4225270920 creator A5013540955 @default.
- W4225270920 creator A5022526821 @default.
- W4225270920 creator A5059767940 @default.
- W4225270920 date "2022-05-23" @default.
- W4225270920 modified "2023-10-18" @default.
- W4225270920 title "Dementia Detection by Fusing Speech and Eye-Tracking Representation" @default.
- W4225270920 cites W1600451534 @default.
- W4225270920 cites W1853705225 @default.
- W4225270920 cites W1971583858 @default.
- W4225270920 cites W2012897754 @default.
- W4225270920 cites W2047509901 @default.
- W4225270920 cites W2074037951 @default.
- W4225270920 cites W2115017507 @default.
- W4225270920 cites W2158516192 @default.
- W4225270920 cites W2239141610 @default.
- W4225270920 cites W2401240531 @default.
- W4225270920 cites W2610961739 @default.
- W4225270920 cites W2775724624 @default.
- W4225270920 cites W2889065492 @default.
- W4225270920 cites W2964051877 @default.
- W4225270920 cites W2972983216 @default.
- W4225270920 cites W3012419921 @default.
- W4225270920 cites W3029634315 @default.
- W4225270920 cites W3037437335 @default.
- W4225270920 cites W3113283374 @default.
- W4225270920 cites W3161417069 @default.
- W4225270920 doi "https://doi.org/10.1109/icassp43922.2022.9747054" @default.
- W4225270920 hasPublicationYear "2022" @default.
- W4225270920 type Work @default.
- W4225270920 citedByCount "3" @default.
- W4225270920 countsByYear W42252709202023 @default.
- W4225270920 crossrefType "proceedings-article" @default.
- W4225270920 hasAuthorship W4225270920A5004434747 @default.
- W4225270920 hasAuthorship W4225270920A5007911614 @default.
- W4225270920 hasAuthorship W4225270920A5013540955 @default.
- W4225270920 hasAuthorship W4225270920A5022526821 @default.
- W4225270920 hasAuthorship W4225270920A5059767940 @default.
- W4225270920 hasConcept C111919701 @default.
- W4225270920 hasConcept C118505674 @default.
- W4225270920 hasConcept C119599485 @default.
- W4225270920 hasConcept C127413603 @default.
- W4225270920 hasConcept C141353440 @default.
- W4225270920 hasConcept C149635348 @default.
- W4225270920 hasConcept C153180895 @default.
- W4225270920 hasConcept C154945302 @default.
- W4225270920 hasConcept C165801399 @default.
- W4225270920 hasConcept C23224414 @default.
- W4225270920 hasConcept C2780513914 @default.
- W4225270920 hasConcept C28490314 @default.
- W4225270920 hasConcept C41008148 @default.
- W4225270920 hasConcept C56461940 @default.
- W4225270920 hasConcept C66322947 @default.
- W4225270920 hasConceptScore W4225270920C111919701 @default.
- W4225270920 hasConceptScore W4225270920C118505674 @default.
- W4225270920 hasConceptScore W4225270920C119599485 @default.
- W4225270920 hasConceptScore W4225270920C127413603 @default.
- W4225270920 hasConceptScore W4225270920C141353440 @default.
- W4225270920 hasConceptScore W4225270920C149635348 @default.
- W4225270920 hasConceptScore W4225270920C153180895 @default.
- W4225270920 hasConceptScore W4225270920C154945302 @default.
- W4225270920 hasConceptScore W4225270920C165801399 @default.
- W4225270920 hasConceptScore W4225270920C23224414 @default.
- W4225270920 hasConceptScore W4225270920C2780513914 @default.
- W4225270920 hasConceptScore W4225270920C28490314 @default.
- W4225270920 hasConceptScore W4225270920C41008148 @default.
- W4225270920 hasConceptScore W4225270920C56461940 @default.
- W4225270920 hasConceptScore W4225270920C66322947 @default.
- W4225270920 hasFunder F4320321543 @default.
- W4225270920 hasFunder F4320330944 @default.
- W4225270920 hasLocation W42252709201 @default.
- W4225270920 hasOpenAccess W4225270920 @default.
- W4225270920 hasPrimaryLocation W42252709201 @default.
- W4225270920 hasRelatedWork W2275988210 @default.
- W4225270920 hasRelatedWork W2357461155 @default.
- W4225270920 hasRelatedWork W2384129116 @default.
- W4225270920 hasRelatedWork W2539985974 @default.
- W4225270920 hasRelatedWork W2552023628 @default.
- W4225270920 hasRelatedWork W2892009249 @default.
- W4225270920 hasRelatedWork W2897067191 @default.
- W4225270920 hasRelatedWork W4224255737 @default.
- W4225270920 hasRelatedWork W4224942917 @default.
- W4225270920 hasRelatedWork W4312658081 @default.
- W4225270920 isParatext "false" @default.
- W4225270920 isRetracted "false" @default.
- W4225270920 workType "article" @default.