Age and gender detection from images with deep learning

Keywords: Deep Learning, image processing, age estimation, gender estimation In our age, where large data is processed at great speeds, deep learning algorithms are used to facilitate the solution of various problems by extracting different parameters from billions of data. In this study, it is aimed to determine the age and sex of the female, male, old, young, child and baby photographs in the existing datasets with deep learning algorithms. In order to realize this estimation algorithm, various deep learning libraries were used and a model was developed with deep learning models and compared with other models. The thesis explains in detail the age and gender determination using deep learning algorithms. The aim of this thesis is to contribute to the literature studies in the field of age and gender determination with deep learning. In the study, firstly, the studies which have performed age and gender estimation by using deep learning have been applied in the literature. The data set used in the application is composed of male and female photographs. Each photo is labeled according to the gender and age of the person. This data set is compiled from wikipedi images, contains 3170 training data and 318 test data. The results of three different models were compared. The study describes in detail the gender prediction using deep learning algorithms and plans to guide the future studies to be carried out.