Authors : H.S. Gunavathi and M. Siddappa
Abstract: Cognitive state recognition is one of the active science researches all around the world and it has grown spontaneously in recent years. However, most research focuses on posed expressions, near frontal recordings and they ignore eye gaze, head pose and considers hand occlusions as noise. It makes tough to tell how the existing methods perform underneath conditions where faces appear in a wide range of poses and occluded by hands. In this study, we propose multimodal approaches for building a real-time cognitive state recognition system in e-Learning environments by integrating hand-over-face gesture with facial expression. Our proposed system performs an average recognition rate of 90.51% with 15.8 fps is robust to variations in facial expressions, hand shapes and occlusions.
H.S. Gunavathi and M. Siddappa, 2019. Automatic Recognition of Cognitive States Using Multimodal Approaches in e-Learning Environments. Journal of Engineering and Applied Sciences, 14: 1286-1294.