BLADERUNNER: Rapid Countermeasure for Synthetic (AI-Generated) StyleGAN Faces

View on arXiv ← Back to list

Authors: Adam Dorian Wong

Published: 2022-10-12 21:05:35+00:00

AI Summary

The research proposes Blade Runner, a rapid countermeasure for detecting StyleGAN-generated synthetic faces. It leverages pre-trained machine learning models to exploit repetitive patterns in StyleGAN images, enabling faster detection than human visual inspection.

Abstract

StyleGAN is the open-sourced TensorFlow implementation made by NVIDIA. It has revolutionized high quality facial image generation. However, this democratization of Artificial Intelligence / Machine Learning (AI/ML) algorithms has enabled hostile threat actors to establish cyber personas or sock-puppet accounts in social media platforms. These ultra-realistic synthetic faces. This report surveys the relevance of AI/ML with respect to Cyber & Information Operations. The proliferation of AI/ML algorithms has led to a rise in DeepFakes and inauthentic social media accounts. Threats are analyzed within the Strategic and Operational Environments. Existing methods of identifying synthetic faces exists, but they rely on human beings to visually scrutinize each photo for inconsistencies. However, through use of the DLIB 68-landmark pre-trained file, it is possible to analyze and detect synthetic faces by exploiting repetitive behaviors in StyleGAN images. Project Blade Runner encompasses two scripts necessary to counter StyleGAN images. Through PapersPlease acting as the analyzer, it is possible to derive indicators-of-attack (IOA) from scraped image samples. These IOAs can be fed back into AmongUs acting as the detector to identify synthetic faces from live operational samples. The opensource copy of Blade Runner may lack additional unit tests and some functionality, but the open-source copy is a redacted version, far leaner, better optimized, and a proof-of-concept for the information security community. The desired end-state will be to incrementally add automation to stay on-par with its closed-source predecessor.

Key findings

Blade Runner shows promise in rapidly detecting StyleGAN-generated images, providing a quicker alternative to human visual inspection. The accuracy is affected by factors like sunglasses and image resolution, highlighting areas for future improvement. The approach relies on readily available pre-trained models, making it accessible to organizations without extensive AI/ML expertise.

Approach

Blade Runner uses two scripts: PapersPlease analyzes a dataset of StyleGAN images to identify consistent patterns in facial landmark coordinates, particularly the eye locations. AmongUs then uses these patterns to detect new StyleGAN images based on eye location.

Datasets

ThisPersonDoesNotExist, GeneratedPhotos, and possibly others implied from references (e.g., FFHQ dataset used to train StyleGAN)

Model(s)

DLIB's 68-landmark pre-trained face landmark detection model and OpenCV for image processing.

Author countries

USA

← Previous