Facial Emotion Recognition Using Deep Learning Models Based on Transfer Learning Techniques with Classifier

Facial Expression Recognition (FER), also known as Facial Emotion Recognition, constitutes an actively discussed subject within the realms of computer vision and machine learning!-- Query ID="Q1" Text="This is to inform you that corresponding author has been identified as per the information available in the Copyright form.." -->.. It extends its influence into numerous disciplines, including education, psychology, human-computer interaction, and marketing research. The efficient recognition of facial expressions holds significant importance in addressing various challenges. This study undertakes a comprehensive exploration of facial emotion detection, employing the FER 2013 dataset. The study involves experimentation with four distinct convolutional neural network architectures: ResNet-V2, MobileNet-V3, Sequential, and Inception-V3. The primary objective is to categorize seven distinct emotions, namely anger, fear, disgust, happiness, surprise, sadness, and neutrality. The outcomes of the experiments conducted on the FER-2013 Dataset reveal that the fine-tuned MobileNet-V3 model outperforms the other methods in terms of performance.