Скачать или смотреть YOLOFeel: Transfer Learning for Real-Time Emotion Recognition in Social Robots

This project aims to deploy facial expression recognition (FER) on the Jetson AGX Orin Developer Kit with the OAK-D S2 camera in real-time to detect the human emotion. The FER dataset was trained on the YOLO11n (You Only Look Once) object detection model using a transfer learning approach to adapt to the data. This project can be integrated into social robots in the future to interact with humans by understanding their emotions, enabling the robot to communicate effortlessly. The six facial expressions recognized are: Angry, Disgust, Fear, Happy, Sad, Surprise, and Neutral. As it can be seen in the above figure, we ere able to detect surprised face.

See our project here:   / yolofeel-transfer-learning-for-real-time-e...