Research Article

Facial Landmark Detection Using Generative Adversarial Network Combined with Autoencoder for Occlusion

Table 2

AUC0. 08 and failure rate of the model key points with different modules on challenging set.

ModelAUC0.08FR (%)

GAN-IA (without SA) + DRN49.975.08
GAN-IA+PFLD52.124.21
GAN-IA+DRN52.683.86
Autoencoder + DRN53.762.94