Detection of Gender and Age using Machine Learning

Abstract

Abstract: Age and gender predictions of unfiltered faces classify unconstrained real-world facial images into predefined ages and gender. Due to its value in intelligent real-world applications, this study topic has undergone significant advancements. Nevertheless, traditionalmethods based on unfiltered benchmarks have proven inefficient at handling large degrees of variation in unconstrained images. Because of their superior performance, Convolutional Neural Networks (CNNs) based approaches have been employed widely in recent years for the classification of jobs, and good quality of performance in facial analysis. In this work,we propose a novel end-to-end CNN approach, to achieve robust age group andgender classification of natural realworld faces. Two-level CNN architecture includesfeature extraction and classification itself. The feature extraction process extracts a feature corresponding to age and gender, and the classification process classifies theface images according to age and gender. Particularly, we address the largevariations in unfiltered real-world faces with a robust image pre-processing algorithm that prepares and processes those facial images before being given into the CNN model.