Research on Static Face Age Estimation Method Based on Deep Learning

In view of the limitations of simply treating age estimation as a classification or regression problem, a fusion model of classification and regression was proposed, and the methods of classification and regression were used at the same time. In view of the influence of other face attributes (gender, race, etc.) on the aging process, attributes such as gender and race are included into the age estimation system, and the correlation information between attributes such as gender and age is fully utilized. In addition, it is planned to use a single multi-task CNN model to complete tasks such as age estimation and gender identification at the same time, so as to reduce the number of models and calculation consumption