ExpressionHop: A Lightweight Human Facial Expression Classifier

Abstract

A lightweight human facial expression recognition (FER) solution aiming at mobile/edge applications is proposed in this work. The solution, called ExpressionHop, consists offour modules: 1) cropping out patches based on facial landmarks, 2) applying filter banks to each patch to generate a rich set of joint spatial-spectral features, 3) conducting the discriminant feature test (DFT) to select features with higher discriminant power, and 4) performing the final classification task with a classifier. We conduct performance benchmarking on ExpressionHop, traditional and deep learning methods on several commonly used FER datasets such as JAFFE, CK + and KDEF. Experimental results show that ExpressionHop achieves comparable or better classification accuracy. Yet, its model size only has 30K parameters, which is significantly lower than those of deep learning methods.