A comparative analysis of hyperparameter effects on CNN architectures for facial emotion recognition

Abstract

This dissertation investigates facial emotion recognition, an area of computer vision that involves identifying human emotions from facial expressions. It approaches facial emotion recognition as a classification task using labelled images from the FER2013 dataset, employing Convolutional Neural Networks for their capacity to process and extract hierarchical features from image data efficiently. This research utilises custom network architectures to conduct a comparative analysis of the impact of various hyperparameters—such as the number of convolutional layers, regularisation parameters, and learning rates—on model performance. Hyperparameters are systematically tuned to determine their effects on accuracy and overall performance. Notably, the best-performing model developed during this research surpassed human-level performance, established as being somewhere between 65% and 68% on the FER2013 dataset according to various studies. These findings provide a foundational understanding of hyperparameter optimisation for facial emotion recognition, demonstrating the impact of different configurations on model performance.