PTW: Pivotal Tuning Watermarking for Pre-Trained Image Generators

by   Nils Lukas, et al.

Deepfakes refer to content synthesized using deep generators, which, when misused, have the potential to erode trust in digital media. Synthesizing high-quality deepfakes requires access to large and complex generators only few entities can train and provide. The threat are malicious users that exploit access to the provided model and generate harmful deepfakes without risking detection. Watermarking makes deepfakes detectable by embedding an identifiable code into the generator that is later extractable from its generated images. We propose Pivotal Tuning Watermarking (PTW), a method for watermarking pre-trained generators (i) three orders of magnitude faster than watermarking from scratch and (ii) without the need for any training data. We improve existing watermarking methods and scale to generators 4 × larger than related work. PTW can embed longer codes than existing methods while better preserving the generator's image quality. We propose rigorous, game-based definitions for robustness and undetectability and our study reveals that watermarking is not robust against an adaptive white-box attacker who has control over the generator's parameters. We propose an adaptive attack that can successfully remove any watermarking with access to only 200 non-watermarked images. Our work challenges the trustworthiness of watermarking for deepfake detection when the parameters of a generator are available.


page 2

page 9


MotionVideoGAN: A Novel Video Generator Based on the Motion Space Learned from Image Pairs

Video generation has achieved rapid progress benefiting from high-qualit...

Transfer learning from language models to image caption generators: Better models may not transfer better

When designing a neural caption generator, a convolutional neural networ...

Vulnerabilities in AI Code Generators: Exploring Targeted Data Poisoning Attacks

In this work, we assess the security of AI code generators via data pois...

Diffusion Guided Domain Adaptation of Image Generators

Can a text-to-image diffusion model be used as a training objective for ...

BROOD: Bilevel and Robust Optimization and Outlier Detection for Efficient Tuning of High-Energy Physics Event Generators

The parameters in Monte Carlo (MC) event generators are tuned on experim...

Data-Free Knowledge Amalgamation via Group-Stack Dual-GAN

Recent advances in deep learning have provided procedures for learning o...

Please sign up or login with your details

Forgot password? Click here to reset