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

RGGID: A Robust and Green GAN-Fake Image Detector

Yao Zhu, Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, USA, yaozhu@usc.edu , Xinyu Wang, Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, USA, Ronald Salloum, School of Computer Science and Engineering, California State University, USA, Hong-Shuo Chen, Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, USA, C.-C. Jay Kuo, Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, USA
 
Suggested Citation
Yao Zhu, Xinyu Wang, Ronald Salloum, Hong-Shuo Chen and C.-C. Jay Kuo (2022), "RGGID: A Robust and Green GAN-Fake Image Detector", APSIPA Transactions on Signal and Information Processing: Vol. 11: No. 2, e38. http://dx.doi.org/10.1561/116.00000005

Publication Date: 28 Dec 2022
© 2022 Y. Zhu, X. Wang, R. Salloum, H-S. Chen and C.-C. J. Kuo
 
Subjects
 
Keywords
Fake image detectiongenerative adversarial networkssubspace learninggreen learningPixelHop
 

Share

Open Access

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

Downloaded: 666 times

In this article:
Introduction 
Related Work 
Proposed RGGID Method 
Experiments 
Analysis 
Conclusion and Future Work 
References 

Abstract

Generative adversarial networks (GANs) are often used to synthesize realistic looking images, which can be a source of dis/misinformation. To detect GAN-fake images effectively, a robust and lightweight detector is proposed and named RGGID (Robust and Green GAN-fake Image Detector) in this work. RGGID is developed under the assumption that GANs fail to generate high-frequency components of real images in high fidelity. Based on this assumption, we design a set of filters using a specific local neighborhood pattern of a pixel, called a PixelHop, and determine the associated discriminant channels. We obtain multiple PixelHops by varying the local patterns, use the validation data to identify discriminant channels, and ensemble their channel responses to yield state-of-the-art detection performance. RGGID offers a green solution since its model size is significantly smaller than that of deep neural networks. Furthermore, we apply common manipulations to real/fake source images, including JPEG compression, resizing and Gaussian additive noise, and demonstrate the robustness of RGGID to these manipulations.

DOI:10.1561/116.00000005

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.