Keyframe based Face in Video Recognition using Multi-Artificial Neural Network

Abstract

Face in video recognition has several challenges. Although deep learning approaches have achieved performance that surpasses people for still image- face recognition, video- face recognition remains a challenging task due to the large volume of data to be processed and intra / intervideo variations in pose, lighting, occlusion, scene, blur, video quality, etc. In this paper, deep convolutional neural network is used for feature extraction and artificial neural network is used for face recognition. The computation overhead of these deep learning approaches is reduced by introducing keyframe based face recognition in video. The low-quality frames are removed by extracting keyframes. The proposed method is tested on YTF dataset and the results are compared with recent methods. The experimental results substantially proved that the proposed method achieves a higher accuracy rate of 98.36% when compared with other recent methods.