GANprintR: Improved Fakes and Evaluation of the State of the Art in Face Manipulation Detection

TL;DR

This paper introduces GANprintR, a method to remove GAN fingerprints from fake facial images, evaluates current detection techniques, and presents a new database to advance face manipulation detection research.

Contribution

The study proposes a novel autoencoder-based approach to remove GAN fingerprints, assesses the robustness of existing detection methods, and introduces the iFakeFaceDB database for benchmarking.

Findings

GANprintR effectively removes GAN fingerprints while preserving image quality.

Current detection systems struggle with unseen spoofing techniques.

Developing robust detection methods remains a significant challenge.

Abstract

The availability of large-scale facial databases, together with the remarkable progresses of deep learning technologies, in particular Generative Adversarial Networks (GANs), have led to the generation of extremely realistic fake facial content, raising obvious concerns about the potential for misuse. Such concerns have fostered the research on manipulation detection methods that, contrary to humans, have already achieved astonishing results in various scenarios. In this study, we focus on the synthesis of entire facial images, which is a specific type of facial manipulation. The main contributions of this study are four-fold: i) a novel strategy to remove GAN "fingerprints" from synthetic fake images based on autoencoders is described, in order to spoof facial manipulation detection systems while keeping the visual quality of the resulting images; ii) an in-depth analysis of the recent literature in facial manipulation detection; iii) a complete experimental assessment of this type of facial manipulation, considering the state-of-the-art fake detection systems (based on holistic deep networks, steganalysis, and local artifacts), remarking how challenging is this task in unconstrained scenarios; and finally iv) we announce a novel public database, named iFakeFaceDB, yielding from the application of our proposed GAN-fingerprint Removal approach (GANprintR) to already very realistic synthetic fake images. The results obtained in our empirical evaluation show that additional efforts are required to develop robust facial manipulation detection systems against unseen conditions and spoof techniques, such as the one proposed in this study.