Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308823043> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4308823043 abstract "With the application of brain-computer interaction technology in various fields, emotion recognition based on EEG signals has been widely studied. However, the existing manually designed methods for extracting emotion features do not have generality, and the deep models have high parameter redundancy and resource overhead, resulting in the inability to perform end-to-end emotion recognition on embedded devices. To address this problem, we propose a knowledge distillation-based residual network emotion recognition method, which first optimizes the residual network applied to image processing to adapt to the EEG emotion classification task, then designs a deep ResNet34 residual network for adaptive extraction of features for EEG emotion classification, and then distills the emotion recognition performance of the deep residual network to the lightweight network ResNet8, which is finally used in SEED-IV and DEAP achieved 83.9% and 81.5% accuracy, respectively, which is 1.1% and 17.7% improvement in accuracy over the model without distillation learning, and verified that the proposed method can achieve model compression and recognition speedup on the basis of close to the classification performance of the deep model for better deployment on embedded devices." @default.
- W4308823043 created "2022-11-16" @default.
- W4308823043 creator A5007051743 @default.
- W4308823043 creator A5042403792 @default.
- W4308823043 creator A5053605791 @default.
- W4308823043 creator A5065660461 @default.
- W4308823043 creator A5070565335 @default.
- W4308823043 creator A5084636770 @default.
- W4308823043 date "2022-10-03" @default.
- W4308823043 modified "2023-10-17" @default.
- W4308823043 title "EEG emotion recognition based on knowledge distillation optimized residual networks" @default.
- W4308823043 cites W2081420711 @default.
- W4308823043 cites W2899077601 @default.
- W4308823043 cites W2963355311 @default.
- W4308823043 cites W3004127093 @default.
- W4308823043 cites W3034223443 @default.
- W4308823043 cites W3034368386 @default.
- W4308823043 cites W3083218890 @default.
- W4308823043 cites W3164439356 @default.
- W4308823043 cites W3202752098 @default.
- W4308823043 doi "https://doi.org/10.1109/iaeac54830.2022.9929699" @default.
- W4308823043 hasPublicationYear "2022" @default.
- W4308823043 type Work @default.
- W4308823043 citedByCount "0" @default.
- W4308823043 crossrefType "proceedings-article" @default.
- W4308823043 hasAuthorship W4308823043A5007051743 @default.
- W4308823043 hasAuthorship W4308823043A5042403792 @default.
- W4308823043 hasAuthorship W4308823043A5053605791 @default.
- W4308823043 hasAuthorship W4308823043A5065660461 @default.
- W4308823043 hasAuthorship W4308823043A5070565335 @default.
- W4308823043 hasAuthorship W4308823043A5084636770 @default.
- W4308823043 hasConcept C108583219 @default.
- W4308823043 hasConcept C111919701 @default.
- W4308823043 hasConcept C11413529 @default.
- W4308823043 hasConcept C119857082 @default.
- W4308823043 hasConcept C152124472 @default.
- W4308823043 hasConcept C153180895 @default.
- W4308823043 hasConcept C154945302 @default.
- W4308823043 hasConcept C155512373 @default.
- W4308823043 hasConcept C178790620 @default.
- W4308823043 hasConcept C185592680 @default.
- W4308823043 hasConcept C204030448 @default.
- W4308823043 hasConcept C28490314 @default.
- W4308823043 hasConcept C41008148 @default.
- W4308823043 hasConcept C52622490 @default.
- W4308823043 hasConceptScore W4308823043C108583219 @default.
- W4308823043 hasConceptScore W4308823043C111919701 @default.
- W4308823043 hasConceptScore W4308823043C11413529 @default.
- W4308823043 hasConceptScore W4308823043C119857082 @default.
- W4308823043 hasConceptScore W4308823043C152124472 @default.
- W4308823043 hasConceptScore W4308823043C153180895 @default.
- W4308823043 hasConceptScore W4308823043C154945302 @default.
- W4308823043 hasConceptScore W4308823043C155512373 @default.
- W4308823043 hasConceptScore W4308823043C178790620 @default.
- W4308823043 hasConceptScore W4308823043C185592680 @default.
- W4308823043 hasConceptScore W4308823043C204030448 @default.
- W4308823043 hasConceptScore W4308823043C28490314 @default.
- W4308823043 hasConceptScore W4308823043C41008148 @default.
- W4308823043 hasConceptScore W4308823043C52622490 @default.
- W4308823043 hasLocation W43088230431 @default.
- W4308823043 hasOpenAccess W4308823043 @default.
- W4308823043 hasPrimaryLocation W43088230431 @default.
- W4308823043 hasRelatedWork W2732542196 @default.
- W4308823043 hasRelatedWork W2773120646 @default.
- W4308823043 hasRelatedWork W2774265021 @default.
- W4308823043 hasRelatedWork W3014300295 @default.
- W4308823043 hasRelatedWork W3186111093 @default.
- W4308823043 hasRelatedWork W4223943233 @default.
- W4308823043 hasRelatedWork W4225161397 @default.
- W4308823043 hasRelatedWork W4309045103 @default.
- W4308823043 hasRelatedWork W4312200629 @default.
- W4308823043 hasRelatedWork W4317987726 @default.
- W4308823043 isParatext "false" @default.
- W4308823043 isRetracted "false" @default.
- W4308823043 workType "article" @default.