Atypical Facial Landmark Localisation with Stacked Hourglass Networks: A Study on 3D Facial Modelling for Medical Diagnosis

TL;DR

This paper investigates the use of stacked hourglass networks for accurate facial landmark detection in atypical 3D facial models, aiming to aid medical diagnosis of facial palsy by identifying distinctive facial features.

Contribution

It evaluates the effectiveness of stacked hourglass networks for landmark localization on atypical faces, demonstrating superior performance over traditional methods.

Findings

Stacked hourglass networks outperform traditional landmark detection methods.

The approach shows promise for medical diagnosis of facial palsy.

Atypical facial features can be effectively modeled using advanced neural networks.

Abstract

While facial biometrics has been widely used for identification purpose, it has recently been researched as medical biometrics for a range of diseases. In this chapter, we investigate the facial landmark detection for atypical 3D facial modelling in facial palsy cases, while potentially such modelling can assist the medical diagnosis using atypical facial features. In our work, a study of landmarks localisation methods such as stacked hourglass networks is conducted and evaluated to ascertain their accuracy when presented with unseen atypical faces. The evaluation highlights that the state-of-the-art stacked hourglass architecture outperforms other traditional methods.