Practical Disruption of Image Translation Deepfake Networks

Authors

  • Nataniel Ruiz Boston University
  • Sarah Adel Bargal Georgetown University
  • Cihang Xie University of California, Santa Cruz
  • Stan Sclaroff Boston University

DOI:

https://doi.org/10.1609/aaai.v37i12.26693

Keywords:

General

Abstract

By harnessing the latest advances in deep learning, image-to-image translation architectures have recently achieved impressive capabilities. Unfortunately, the growing representational power of these architectures has prominent unethical uses. Among these, the threats of (1) face manipulation ("DeepFakes") used for misinformation or pornographic use (2) "DeepNude" manipulations of body images to remove clothes from individuals, etc. Several works tackle the task of disrupting such image translation networks by inserting imperceptible adversarial attacks into the input image. Nevertheless, these works have limitations that may result in disruptions that are not practical in the real world. Specifically, most works generate disruptions in a white-box scenario, assuming perfect knowledge about the image translation network. The few remaining works that assume a black-box scenario require a large number of queries to successfully disrupt the adversary's image translation network. In this work we propose Leaking Transferable Perturbations (LTP), an algorithm that significantly reduces the number of queries needed to disrupt an image translation network by dynamically re-purposing previous disruptions into new query efficient disruptions.

Downloads

Published

2023-06-26

How to Cite

Ruiz, N., Bargal, S. A., Xie, C., & Sclaroff, S. (2023). Practical Disruption of Image Translation Deepfake Networks. Proceedings of the AAAI Conference on Artificial Intelligence, 37(12), 14478-14486. https://doi.org/10.1609/aaai.v37i12.26693

Issue

Section

AAAI Special Track on AI for Social Impact