Visual-to-EEG cross-modal knowledge distillation for continuous emotion recognition
作者：
Highlights：
• By taking the above visual and EEG models as the teacher and student, we develop a cross-modal knowledge distillation method to improve the EEG-based continuous emotion recognition using visual knowledge.
• The standalone version of teacher and student without knowledge distillation can outperform baseline.
• The student model taught by the labels and the visual knowledge produces results with statistical significance against its counterpart without knowledge distillation.
• To the best of the authors’ knowledge, this is the first work on visual-to-EEG cross-modal knowledge distillation for continuous emotion recognition.
• The code is to be publicly available.
摘要
•By taking the above visual and EEG models as the teacher and student, we develop a cross-modal knowledge distillation method to improve the EEG-based continuous emotion recognition using visual knowledge.•The standalone version of teacher and student without knowledge distillation can outperform baseline.•The student model taught by the labels and the visual knowledge produces results with statistical significance against its counterpart without knowledge distillation.•To the best of the authors’ knowledge, this is the first work on visual-to-EEG cross-modal knowledge distillation for continuous emotion recognition.•The code is to be publicly available.
论文关键词：Continuous emotion recognition,Knowledge distillation,Cross-modality
论文评审过程：Received 26 August 2021, Revised 3 May 2022, Accepted 3 June 2022, Available online 3 June 2022, Version of Record 7 June 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108833
