Face-Based Engagement Detection Methods: A Review

The detection of student engagement in online learning environments has become increasingly important with the widespread adoption of e-learning platforms. This paper reviews current approaches for monitoring student engagement based on facial expressions, gaze tracking, fatigue and drowsiness detection, and multimodal systems. By analyzing facial expressions, systems can detect emotional states such as happiness, frustration, and boredom, offering real-time feedback to instructors. Gaze tracking provides insights into students focus, although challenges such as hardware costs and lighting conditions affect its accuracy. Fatigue and drowsiness detection, through blinking and yawning analysis, helps identify cognitive overload, while multimodal systems that combine facial, behavioral, and physiological data offer a more comprehensive picture of engagement. This review highlights the potential of these methods while addressing the need for more robust, scalable, and privacy-conscious systems for real-time engagement monitoring in di-verse e-learning contexts.