Scoring facial attractiveness with deep convolutional neural networks: How training on standardized images reduces the bias of facial expressions.

Journal: Orthodontics & craniofacial research
PMID: 38825845

Abstract

OBJECTIVE: In many medical disciplines, facial attractiveness is part of the diagnosis, yet its scoring might be confounded by facial expressions. The intent was to apply deep convolutional neural networks (CNN) to identify how facial expressions affect facial attractiveness and to explore whether a dedicated training of the CNN is able to reduce the bias of facial expressions.

Authors

  • Dorothea Obwegeser
    Clinic of Orthodontics and Pediatric Dentistry, Center of Dental Medicine, University of Zurich, Switzerland.
  • Radu Timofte
    Computer Vision Laboratory, D-ITET, ETH Zurich, Switzerland.
  • Christoph Mayer
    Computer Vision Laboratory, Department of Information Technology and Electrical Engineering, ETH Zurich, Switzerland.
  • Michael M Bornstein
    Department of Oral Health & Medicine, University Center for Dental Medicine Basel UZB, University of Basel, Basel, Switzerland.
  • Marc A Schätzle
    Clinic of Orthodontics and Pediatric Dentistry, Center of Dental Medicine, University of Zurich, Switzerland.
  • Raphael Patcas
    Clinic of Orthodontics and Pediatric Dentistry, Center of Dental Medicine, University of Zurich, Switzerland.

Keywords

External Resources

Popular Topics
Recent Journals