Age Prediction from Facial Images Using Deep Learning Architecture

Predicting age and gender through images is a common computer vision problem with many practical applications. However, this problem faces many difficulties because a person's age can be affected by genetics, living environment, diet, health, gender, and other factors. Therefore, the accuracy of the prediction model may decrease due to the enormous diversity and variability in the data. In this study, we use three models, including Unet, MobileNets, and EfficientNets, to test the performance of predicting a person's age and gender through their photos. In addition, we also adjust the learning rate parameter to find optimal performance. The best results for gender prediction are achieved by the Unet model with the highest accuracy of 97.22 %, and the MobileNets model gives age prediction results with MAE = 2.248, learning rate 0.001 for optimal performance in the models of our study.