This article investigates the use of facial recognition technology to identify Roman emperors from their portraits. Current methods rely on coiffure patterns, which can be unreliable. The authors utilize transfer learning with a pre-trained InceptionResnet-V1 model, augmenting it with images of Roman imperial portraits. Results show that the model accurately classifies most images, suggesting facial recognition offers a new, empirical approach to identifying Roman emperors.
Publisher
Humanities and Social Sciences Communications
Published On
Mar 14, 2022
Authors
Darshan Srirangachar Ramesh, Sam Heijnen, Olivier Hekster, Luuk Spreeuwers, Florens de Wit
Tags
facial recognition
Roman emperors
portrait identification
transfer learning
InceptionResnet-V1
artificial intelligence
historical analysis
Related Publications
Explore these studies to deepen your understanding of the subject.
