Adversarially robust deepfake media detection using fused convolutional neural network predictions

Authors

Sohail Ahmed Khan, CYENS Centre of Excellence & Mohamed bin Zayed University of Artificial IntelligenceFollow
Alessandro Artusi, CYENS Centre of Excellence
Hang Dai, Mohamed bin Zayed University of Artificial IntelligenceFollow

Document Type

Article

Publication Title

arXiv

Abstract

Deepfakes are synthetically generated images, videos or audios, which fraudsters use to manipulate legitimate information. Current deepfake detection systems struggle against unseen data. To address this, we employ three different deep Convolutional Neural Network (CNN) models, (1) VGG16, (2) InceptionV3, and (3) XceptionNet to classify fake and real images extracted from videos. We also constructed a fusion of the deep CNN models to improve the robustness and generalisation capability. The proposed technique outperforms state-of-the-art models with 96.5% accuracy, when tested on publicly available DeepFake Detection Challenge (DFDC) test data, comprising of 400 videos. The fusion model achieves 99% accuracy on lower quality DeepFakeTIMIT dataset videos and 91.88% on higher quality DeepFakeTIMIT videos. In addition to this, we prove that prediction fusion is more robust against adversarial attacks. If one model is compromised by an adversarial attack, the prediction fusion does not let it affect the overall classification. © 2021, CC BY.

DOI

doi.org/10.48550/arXiv.2102.05950

Publication Date

2-11-2021

Keywords

Computer Vision and Pattern Recognition (cs.CV), Cryptography and Security (cs.CR)

Comments

Preprint: arXiv

Recommended Citation

S.A. Khan, Artusi A., and H. Dai, "Adversarially robust deepfake media detection using fused convolutional neural network predictions", 2021, arXiv:2102.05950