Comparison of Deep Learning Architectures for Facial Recognition

Abstrak

Facial recognition technology in modern security systems, including access control, identity verification, and digital devices. This study aims to compare the performance of three widely used deep learning architectures, Convolutional Neural Network (CNN), VGG16, and FaceNet512, in processing and identifying facial features. A quantitative approach was employed through computational experiments using facial image datasets. The performance of each model was evaluated using accuracy, precision, recall, and F1 score to assess its effectiveness in facial recognition tasks. The study revealed significant differences in the performance of each architecture, both in terms of recognition accuracy and processing efficiency. CNN, VGG16, and FaceNet512 each demonstrated distinct strengths and limitations. These findings provide valuable insights for selecting the most suitable deep learning architecture for practical and academic applications in facial biometric security systems.