Human Emotion Recognition Using ResNet Architechture

Emotion detection plays a crucial role in enabling systems to accurately interpret and respond to human emotions, thereby enhancing human-computer interaction. This re- search leverages the Residual Neural Network (ResNet) architecture—a deep learning model specifically designed to tackle challenges like the vanishing gradient problem in deep networks—to deliver an improved approach to emotion detection. By leveraging ResNet’s ability to learn residuals, the proposed system achieves superior accuracy in classifying emotions from facial expressions, outperforming traditional models. Com- pared to KNearest Neighbors (KNN), which struggles with high-dimensional data, and Convolutional Neural Networks (CNNs), which require large datasets and computational resources, ResNet excels with its residual connections, allowing deeper networks to ef- ficiently learn subtle facial features. This leads to better performance in challenging conditions like lighting variations and occlusions. Despite its higher computational cost, ResNet’s accuracy makes it the ideal choice for emotion detection and face recognition in this study.