Facial Expression Recognition Based on Neural Network and Feature Extraction

Abstract

Facial expression recognition has always been an interesting and challenging problem, which is easily affected by various factors in practical applications, such as illumination, posture, facial occlusion, age, race and so on. This paper focuses on facial expression recognition based on neural network and feature extraction. In this paper, a new convolutional neural network model for facial expression recognition is proposed to solve the problem that the traditional convolutional neural network is too complex. In the initial stage, the interleaved group convolution structure is adopted, which effectively reduces the number of channel connections between network layers, and at the same time, the interleaved structure also enables effective information communication between different volume units. Through verification and comparison on CK+ and FER datasets in this paper, the improved convolutional neural network facial expression recognition model proposed in this paper can better complete the task of facial expression recognition.