Deep Learning-Based Face Detection, Feature Extraction, and Face Recognition from Video: A Comprehensive Review

Abstract

Face recognition has become one of the most prominent biometric technologies due to its extensive applications in surveillance, access control, authentication, human-computer interaction, and intelligent security systems. Recent advances in deep learning, particularly Convolutional Neural Networks (CNNs), have significantly improved the accuracy and robustness of face detection, feature extraction, and face recognition under challenging real-world conditions. This paper presents a comprehensive review of deep learning-based techniques employed across the complete face recognition pipeline. A comparative analysis of existing studies is presented to highlight the evolution of deep learning techniques and their effectiveness in improving recognition accuracy and computational efficiency. The review also discusses widely used benchmark datasets, performance evaluation metrics, and the major challenges encountered in unconstrained environments, such as pose variation, illumination changes, occlusion, facial expressions, aging, and low-resolution imagery. Finally, emerging research directions, including lightweight deep learning models, attention mechanisms, Vision Transformers, self-supervised learning, explainable artificial intelligence, and real-time video-based face recognition, are outlined to provide insights for future research. This review serves as a comprehensive reference for researchers and practitioners seeking a thorough understanding of recent developments and future trends in deep learning-based face recognition.