Emotion Recognition on FER-2013 Face Images Using Fine-Tuned VGG-16
Gede Putra Kusuma, Jonathan Jonathan, Andreas Pangestu Lim · Advances in Science Technology and Engineering Systems Journal · 2020
Facial emotion recognition is one among many popular and challenging tasks in the field of computer vision. Numerous researches have been conducted on this task and each proposed either standalone-or ensemble-based processing technique. While many researches strive for better accuracy, this research also attempts to increase the processing efficiency of computer correctly classifying human emotions based on human face by utilizing a single standalone-based neural network. This research proposes the use of standalone-based modified Convolutional Neural Network (CNN) based on VisualGeometry Group -16 (VGG-16) classification model which was pretrained on ImageNet dataset and fine-tuned for emotion classification.The classification is performed on the publicly available FER-2013 dataset of over 35,000 face images with in-the-wild setting for 7 distinct emotions with the provided 80% training, 10% validation, and 10% testing data distributions.The proposed approach outperforms most standalone-based model results with 69.40% accuracy.