Smart Emotion and Attention Monitoring System for Students using Real-Time Facial Emotion Recognition (FER)

Education is evolving rapidly with the integration of Artificial Intelligence (AI) and data-driven teaching methodolo- gies. However, teachers often find it difficult to assess students’ emotional states and attention levels during lectures. This paper presents a Smart Emotion and Attention Monitoring System that leverages Facial Emotion Recognition (FER) and Deep Learning to analyze students’ emotions in real-time. Using Convolutional Neural Networks (CNNs) trained on benchmark datasets such as FER2013 and fine-tuned on classroom-like samples, the system classifies emotions into multiple categories and computes engagement metrics for teachers and parents. Privacy-preserving measures and ethical guidelines are integrated into the design. Experimental results on simulated classroom data indicate promising accuracy and low-latency behavior suitable for near-real-time feedback.