Human Emotions Analysis and Recognition Using EEG Signals in Response to 360$^\circ$ Videos
Haseeb ur Rahman Abbasi, Zeeshan Rashid, Muhammad Majid, Syed Muhammad Anwar
TL;DR
This work tackles EEG-based emotion recognition in response to immersive VR 360° videos, addressing the scarcity of VR-elicited physiological datasets. It proposes a four-stage pipeline—data acquisition, preprocessing, feature extraction, and classification—using a Muse EEG headband and Oculus Quest 2 to collect data from 40 participants across four emotion classes, with 34 frequency-domain features derived from five bands. An SVM classifier with multiple kernels shows that the polynomial kernel yields the best performance, achieving a cross-validated average accuracy of $85.54\%$ (max $90.20\%$) and a test accuracy of $82.03\%$ for four-class emotion recognition. The study contributes a VR-based EEG dataset and demonstrates the viability of emotion recognition in VR contexts, with implications for VR-HCI applications and potential multimodal emotion analysis in future work.
Abstract
Emotion recognition (ER) technology is an integral part for developing innovative applications such as drowsiness detection and health monitoring that plays a pivotal role in contemporary society. This study delves into ER using electroencephalography (EEG), within immersive virtual reality (VR) environments. There are four main stages in our proposed methodology including data acquisition, pre-processing, feature extraction, and emotion classification. Acknowledging the limitations of existing 2D datasets, we introduce a groundbreaking 3D VR dataset to elevate the precision of emotion elicitation. Leveraging the Interaxon Muse headband for EEG recording and Oculus Quest 2 for VR stimuli, we meticulously recorded data from 40 participants, prioritizing subjects without reported mental illnesses. Pre-processing entails rigorous cleaning, uniform truncation, and the application of a Savitzky-Golay filter to the EEG data. Feature extraction encompasses a comprehensive analysis of metrics such as power spectral density, correlation, rational and divisional asymmetry, and power spectrum. To ensure the robustness of our model, we employed a 10-fold cross-validation, revealing an average validation accuracy of 85.54\%, with a noteworthy maximum accuracy of 90.20\% in the best fold. Subsequently, the trained model demonstrated a commendable test accuracy of 82.03\%, promising favorable outcomes.