DeepFakes: a New Threat to Face Recognition? Assessment and Detection

TL;DR

This paper evaluates the vulnerability of face recognition systems to GAN-generated Deepfake videos and assesses detection methods, highlighting the need for improved detection techniques due to increasing Deepfake quality.

Contribution

It provides the first publicly available Deepfake video dataset, analyzes the impact of video quality on face recognition vulnerability, and evaluates baseline detection approaches.

Findings

Face recognition systems have high false acceptance rates with Deepfakes.

Audio-visual detection methods are ineffective against Deepfakes.

Visual quality metrics show promise for Deepfake detection.

Abstract

It is becoming increasingly easy to automatically replace a face of one person in a video with the face of another person by using a pre-trained generative adversarial network (GAN). Recent public scandals, e.g., the faces of celebrities being swapped onto pornographic videos, call for automated ways to detect these Deepfake videos. To help developing such methods, in this paper, we present the first publicly available set of Deepfake videos generated from videos of VidTIMIT database. We used open source software based on GANs to create the Deepfakes, and we emphasize that training and blending parameters can significantly impact the quality of the resulted videos. To demonstrate this impact, we generated videos with low and high visual quality (320 videos each) using differently tuned parameter sets. We showed that the state of the art face recognition systems based on VGG and Facenet neural networks are vulnerable to Deepfake videos, with 85.62% and 95.00% false acceptance rates respectively, which means methods for detecting Deepfake videos are necessary. By considering several baseline approaches, we found that audio-visual approach based on lip-sync inconsistency detection was not able to distinguish Deepfake videos. The best performing method, which is based on visual quality metrics and is often used in presentation attack detection domain, resulted in 8.97% equal error rate on high quality Deepfakes. Our experiments demonstrate that GAN-generated Deepfake videos are challenging for both face recognition systems and existing detection methods, and the further development of face swapping technology will make it even more so.