APSIPA Transactions on Signal and Information Processing > Vol 11 > Issue 2

DefakeHop++: An Enhanced Lightweight Deepfake Detector

Hong-Shuo Chen, University of Southern California, USA, hongshuo@usc.edu , Shuowen Hu, DEVCOM Army Research Laboratory, USA, Suya You, DEVCOM Army Research Laboratory, USA, C.-C. Jay Kuo, University of Southern California, USA
 
Suggested Citation
Hong-Shuo Chen, Shuowen Hu, Suya You and C.-C. Jay Kuo (2022), "DefakeHop++: An Enhanced Lightweight Deepfake Detector", APSIPA Transactions on Signal and Information Processing: Vol. 11: No. 2, e41. http://dx.doi.org/10.1561/116.00000126

Publication Date: 28 Dec 2022
© 2022 H.-S. Chen, S. Hu, S. You and C.-C. J. Kuo
 
Subjects
 
Keywords
Deepfake detectiongreen learninggreen AIweakly-supervised learning
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 980 times

In this article:
Introduction 
Review of Related Work 
DefakeHop++ 
Experiments 
Conclusion and Future Work 
References 

Abstract

On the basis of DefakeHop, an enhanced lightweight Deepfake detector called DefakeHop++ is proposed in this work. The improvements lie in two areas. First, DefakeHop examines three facial regions (i.e., two eyes and mouth) while DefakeHop++ includes eight more landmarks for broader coverage. Second, for discriminant features selection, DefakeHop uses an unsupervised approach while DefakeHop++ adopts a more effective approach with supervision, called the Discriminant Feature Test (DFT). In DefakeHop++, rich spatial and spectral features are first derived from facial regions and landmarks automatically. Then, DFT is used to select a subset of discriminant features for classifier training. As compared with MobileNet v3 (a lightweight CNN model of 1.5M parameters targeting at mobile applications), DefakeHop++ has a model of 238K parameters, which is 16% of MobileNet v3. Furthermore, DefakeHop++ outperforms MobileNet v3 in Deepfake image detection performance in a weakly-supervised setting.

DOI:10.1561/116.00000126

Companion

APSIPA Transactions on Signal and Information Processing Special Issue - Multi-Disciplinary Dis/Misinformation Analysis and Countermeasures: Articles Overview
See the other articles that are part of this special issue.