Published Paper Details:

Facial expressions serve as a powerful medium of non-verbal communication, offering insights into human emotions. This paper presents a real-time facial emotion detection system leveraging Convolutional Neural Networks (CNNs) for accurate emotion classification and Haar Cascade Classifiers for efficient face detection. The system is trained on the FER-2013 dataset, which includes seven emotion categories: angry, disgust, fear, happy, neutral, sad, and surprise. By preprocessing facial images into grayscale and resizing them to 48x48 pixels, the model ensures consistent input quality. The CNN architecture extracts intricate features from facial images, enabling robust emotion recognition. The system integrates real-time webcam input, allowing continuous emotion detection and overlaying the predicted emotion on the video feed. Experimental results demonstrate the system's ability to achieve high accuracy on both static and dynamic datasets, with minimal latency during real-time processing. Applications of this work span diverse fields, including mental health monitoring, human-computer interaction, and adaptive learning systems. Future enhancements include expanding the dataset, optimizing for edge devices, and incorporating multimodal data for improved accuracy and adaptability.