Abstract
Machine learning (ML) is one of the fastest-developing topics today, straddling the boundary between statistics and computer science, as well as data science. It is a type of artificial intelligence that allows software applications to become more accurate at predicting outcomes without being explicitly programmed. And It addresses the difficulty of the way to assemble gadgets that enhance themselves via experience, and make conclusions with minimum human assistance. For this purpose, there arises a need to use various statistical methods of face recognition’ models, such as (DeepFace) and (OpenFace). DeepFace is the most lightweight face recognition and facial attribute analysis library for Python, and is currently on the verge of human-level precision. OpenFace on the other hand is an open source deep learning facial recognition model based on Google's Facenet model. In this paper, we will discuss the face recognition comparison between two models DeepFace and OpenFace on the calibrators of (Accuracy, Error Rate and Verification Time). DeepFace showed a higher accuracy rate by (3%) than that of OpenFace, and a lower error rate by (3%). Whereas OpenFace delivered with a minimum time shorter than that of DeepFace by (0.061323) second.