DAHMEN IMEN

Development of an intelligent solution for emotional and behavioral analysis to evaluate student engagement in the classroom, using three complementary models: Emotion Model using MobileNetV2 on the RAF-DB dataset, achieving 84% accuracy. Engagement Model combining MobileNetV2 with a GRU module and Attention mechanism, trained on the DAISEE dataset, achieving 75% accuracy. Behavioral Model using YOLOv11s on a multi-source dataset for action detection, achieving 72% mAP50. Technologies : PyTorch, OpenCV, Python

Development of a deep learning model capable of detecting and classifying facial emotions in real-time. Using MobileNetv2 on the RAF-DB dataset, achieving 79% accuracy Technologies: TensorFlow, OpenCV