Age-Invariant Facial Expression Classification Method Using Deep Learning

This paper proposes a method for facial expression classification as a solution to some of the challenges caused by age variations in image classification tasks. The proposed method utilizes deep learning neural networks pre-trained for image classification tasks which are then fine-tuned using a training dataset composed of mixed age facial expression images. The experimental results obtained after testing the model trained on a mixed age dataset, and another model trained on a standard dataset, using new images from a mixed age test dataset, show that the proposed model has better performance at facial expression classification task involving age variations in the input images.