Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Data

View on arXiv ← Back to list

Authors: Ning Yu, Vladislav Skripniuk, Sahar Abdelnabi, Mario Fritz

Published: 2020-07-16 16:49:55+00:00

AI Summary

This paper proposes a proactive deepfake detection method by embedding artificial fingerprints into generative model training data. These fingerprints transfer to the generated images, enabling easy detection and attribution of deepfakes, regardless of the generative model used.

Abstract

Photorealistic image generation has reached a new level of quality due to the breakthroughs of generative adversarial networks (GANs). Yet, the dark side of such deepfakes, the malicious use of generated media, raises concerns about visual misinformation. While existing research work on deepfake detection demonstrates high accuracy, it is subject to advances in generation techniques and adversarial iterations on detection countermeasure techniques. Thus, we seek a proactive and sustainable solution on deepfake detection, that is agnostic to the evolution of generative models, by introducing artificial fingerprints into the models. Our approach is simple and effective. We first embed artificial fingerprints into training data, then validate a surprising discovery on the transferability of such fingerprints from training data to generative models, which in turn appears in the generated deepfakes. Experiments show that our fingerprinting solution (1) holds for a variety of cutting-edge generative models, (2) leads to a negligible side effect on generation quality, (3) stays robust against image-level and model-level perturbations, (4) stays hard to be detected by adversaries, and (5) converts deepfake detection and attribution into trivial tasks and outperforms the recent state-of-the-art baselines. Our solution closes the responsibility loop between publishing pre-trained generative model inventions and their possible misuses, which makes it independent of the current arms race. Code and models are available at https://github.com/ningyu1991/ArtificialGANFingerprints .

Key findings

The proposed fingerprinting method achieves nearly perfect deepfake detection and attribution accuracy across various generative models and datasets. It outperforms state-of-the-art baselines and is robust to image and model perturbations. The fingerprints are also difficult for adversaries to detect.

Approach

The authors embed artificial fingerprints into training data using a deep learning-based steganography technique. These fingerprints surprisingly transfer to the generated images, allowing for detection and attribution by decoding the fingerprints from the generated images.

Datasets

CelebA, LSUN Bedroom, LSUN Cat, CIFAR-10, Horse→Zebra, Cat→Dog

Model(s)

ProGAN, StyleGAN, StyleGAN2, BigGAN, CUT. A steganography encoder and decoder are also used.

Author countries

USA, Germany

← Previous