Video-based age and gender recognition in mobile applications
Angelina Kharchevnikova, Andrey Savchenko · 2018
In this paper we develop the age and gender recognition mobile system using deep convolutional neural networks for mobile applications. The brief literature survey on the age/gender problem in retail applications is presented. The comparative analysis of classifier fusion algorithms to aggregate decisions for individual frames is provided. In order to improve the age and gender identification accuracy we implement the video-based recognition system with several aggregation methods. We provide the experimental comparison for IJB-A, Indian Movies, Kinect and EmotiW2018 datasets. It is demonstrated that the most accurate decisions are obtained using the geometric mean and mathematical expectation of the outputs at softmax layers of the convolutional neural networks for gender recognition and age prediction, respectively. As a result, the off-line application of the proposed system is implemented on the Android platform.