# Ulixes: Facial Recognition Privacy with Adversarial Machine Learning

arXiv: https://arxiv.org/abs/2010.10242 · HTML: https://mlsanity.com/paper/2010.10242/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Oct 2020.

## Built on by (most cited first, 7 of 7)

- [SoK: Anti-Facial Recognition Technology](https://mlsanity.com/paper/2112.04558/) (2021)
- [Toward Face Biometric De-identification using Adversarial Examples](https://mlsanity.com/paper/2302.03657/) (2023)
- [Adversarial Machine Learning for Social Good: Reframing the Adversary as an Ally](https://mlsanity.com/paper/2310.03614/) (2023)
- [Privacy Threats in Stable Diffusion Models](https://mlsanity.com/paper/2311.09355/) (2023)
- [Focused Adversarial Attacks](https://mlsanity.com/paper/2205.09624/) (2022)
- [Group-based Robustness: A General Framework for Customized Robustness in the Real World](https://mlsanity.com/paper/2306.16614/) (2023)
- [Scale Invariant Privacy Preserving Video via Wavelet Decomposition](https://mlsanity.com/paper/2211.03690/) (2022)

## Abstract (arXiv)

Facial recognition tools are becoming exceptionally accurate in identifying people from images. However, this comes at the cost of privacy for users of online services with photo management (e.g. social media platforms). Particularly troubling is the ability to leverage unsupervised learning to recognize faces even when the user has not labeled their images. In this paper we propose Ulixes, a strategy to generate visually non-invasive facial noise masks that yield adversarial examples, preventing the formation of identifiable user clusters in the embedding space of facial encoders. This is applicable even when a user is unmasked and labeled images are available online. We demonstrate the effectiveness of Ulixes by showing that various classification and clustering methods cannot reliably label the adversarial examples we generate. We also study the effects of Ulixes in various black-box settings and compare it to the current state of the art in adversarial machine learning. Finally, we challenge the effectiveness of Ulixes against adversarially trained models and show that it is robust to countermeasures.