Responsible Disclosure of Generative Models Using Scalable Fingerprinting
arXiv:2012.08726
Abstract
Over the past years, deep generative models have achieved a new level of performance. Generated data has become difficult, if not impossible, to be distinguished from real data. While there are plenty of use cases that benefit from this technology, there are also strong concerns on how this new technology can be misused to generate deep fakes and enable misinformation at scale. Unfortunately, current deep fake detection methods are not sustainable, as the gap between real and fake continues to close. In contrast, our work enables a responsible disclosure of such state-of-the-art generative models, that allows model inventors to fingerprint their models, so that the generated samples containing a fingerprint can be accurately detected and attributed to a source. Our technique achieves this by an efficient and scalable ad-hoc generation of a large population of models with distinct fingerprints. Our recommended operation point uses a 128-bit fingerprint which in principle results in more than identifiable models. Experiments show that our method fulfills key properties of a fingerprinting mechanism and achieves effectiveness in deep fake detection and attribution. Code and models are available at https://github.com/ningyu1991/ScalableGANFingerprints .
arXiv commentAccepted to ICLR'22 as Spotlight
References in corpus (12)
- Auto-Encoding Variational Bayes
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Large Scale GAN Training for High Fidelity Natural Image Synthesis
- Spectral Normalization for Generative Adversarial Networks
- MesoNet: a Compact Facial Video Forgery Detection Network
- Embedding Watermarks into Deep Neural Networks
- VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
- Generating Steganographic Images via Adversarial Training
- Unmasking DeepFakes with simple Features
- Amortised MAP Inference for Image Super-resolution
- Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis
- Implicit Maximum Likelihood Estimation