A Comparative Study Between Deep Learning and Traditional Machine Learning Techniques for Facial Biometric Recognition

Abstract

There is a growing incentive to use biometric technology to improve and even replace traditional security methods. Biometric modalities are characteristics drawn from the human body, which are unique to each individual and can be used to establish their identity in a population. Among the biometric modalities, the face is the most commonly seen and used in daily life. Several works have been proposed involving Deep Learning, with emphasis on the Convolutional Neural Networks, for facial recognition. However none of these studies perform a detailed comparative study between traditional machine learning techniques and Deep Learning presenting the pros and cons of each one. In this context, the present work aims to conduct a comparative study between traditional machine learning techniques, such as K-Nearest Neighbors, Optimum-Path Forest, Support Vector Machine, Extreme Learning Machine, Artificial Neural Networks and Deep Learning, focusing on Convolutional Neural Networks, for facial recognition.