Exposing GAN-generated Faces Using Inconsistent Corneal Specular Highlights

by   Shu Hu, et al.

Sophisticated generative adversary network (GAN) models are now able to synthesize highly realistic human faces that are difficult to discern from real ones visually. GAN synthesized faces have become a new form of online disinformation. In this work, we show that GAN synthesized faces can be exposed with the inconsistent corneal specular highlights between two eyes. We show that such artifacts exist widely and further describe a method to extract and compare corneal specular highlights from two eyes. Qualitative and quantitative evaluations of our method suggest its simplicity and effectiveness in distinguishing GAN synthesized faces.


page 1

page 2

page 3

page 4

page 6


Exposing GAN-synthesized Faces Using Landmark Locations

Generative adversary networks (GANs) have recently led to highly realist...

Stereotype-Free Classification of Fictitious Faces

Equal Opportunity and Fairness are receiving increasing attention in art...

Annotating shadows, highlights and faces: the contribution of a 'human in the loop' for digital art history

While automatic computational techniques appear to reveal novel insights...

Defending against GAN-based Deepfake Attacks via Transformation-aware Adversarial Faces

Deepfake represents a category of face-swapping attacks that leverage ma...

Fighting deepfakes by detecting GAN DCT anomalies

Synthetic multimedia content created through AI technologies, such as Ge...

TableGAN-MCA: Evaluating Membership Collisions of GAN-Synthesized Tabular Data Releasing

Generative Adversarial Networks (GAN)-synthesized table publishing lets ...

Ensembling with Deep Generative Views

Recent generative models can synthesize "views" of artificial images tha...