Abstract
Machine recognition of human emotions has attracted more and more attention for its wide application in recent years. As a spontaneous signal of human behavior, eye gaze is utilized for emotion recognition. Compared with electroencephalogram (EEG) signals, eye gaze data is more available, and it can be used in many practical applications, such as virtual reality (VR). In this paper, a new method of human emotion recognition based on eye gaze is proposed. Firstly, a new set of eye gaze features is proposed which consists of eye gaze sequences, extracted statistical feature sequences and spectral feature sequences. Then, the eye gaze feature set is input into the convolutional layer to extract high-level features. Finally, these high-level features and the eye gaze feature set are combined to complete the mapping of features to human emotions. Experiments on MAHNOB-HCI dataset demonstrate the effectiveness of this method.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
