Gender and Age Estimation from Human Faces Based on Deep Learning Techniques: A Review

Abstract

Due to the advancement of the methodologies employed in this field and the increased attention being paid to the use of deep learning techniques, particularly convolutional neural networks, gender and age estimates have recently assumed a significant amount of relevance (CNNs). It is important to accurately estimate a person's gender and age since it is used in many applications for smart devices, including those related to security, health, and other areas. Although there have been several studies and research in this area, gender and age estimation still confronts certain problems and difficulties, such as existing of earrings, races, masked faces, makeup, etc. which might interfere with the systems' operations and decrease their accuracy. In this paper, we assess the accuracy of the models employed in three of the most well-known datasets: MORPH2, FG-NET, and OUI-Adience. Our focus is on the best and most recent technology available in this field. Additionally, we have mentioned a list of most of the challenges that may face the process of estimating age and gender, as well as a list of applications and areas in which it can be used.