Venue: ACM International Conference on Multimedia

Authors: Yucheng Liu, Ziyu Jia, Haichao Wang

We propose EmotionKD, a framework for cross-modal knowledge distillation that simultaneously models the heterogeneity and interactivity of GSR and EEG signals. By using knowledge distillation, fully fused multi-modal features can be transferred to an unimodal GSR model. An adaptive feedback mechanism enables the multi-modal model to dynamically adjust according to the performance of the unimodal model.

Tags: Knowledge Distillation, Emotion Recognition, Physiological Signals