Published Paper Details:

Facial Emotion Detection is an important application of computer vision and deep learning that enables machines to recognize human emotions from facial expressions. This paper presents an efficient real-time facial emotion detection system using the ResNet-18 deep convolutional neural network architecture trained on the Extended Cohn-Kanade (CK+) dataset. The proposed system performs face detection using OpenCV Haar Cascade and applies preprocessing techniques such as resizing, normalization, and grayscale conversion before feature extraction. The ResNet-18 model automatically learns discriminative facial features and classifies expressions into seven emotion categories: happy, sad, angry, fear, surprise, disgust, and neutral. The system is capable of predicting emotions from live webcam input with high accuracy and confidence levels. Experimental results demonstrate reliable performance in real-time environments, making the system suitable for applications such as human-computer interaction, smart classrooms, healthcare monitoring, and driver assistance systems. The proposed approach improves classification efficiency while maintaining computational simplicity for practical deployment.