# Gender Privacy: An Ensemble of Semi Adversarial Networks for Confounding Arbitrary Gender Classifiers

arXiv: https://arxiv.org/abs/1807.11936 · HTML: https://mlsanity.com/paper/1807.11936/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2018.

## Built on by (most cited first, 10 of 10)

- [Meta Balanced Network for Fair Face Recognition](https://mlsanity.com/paper/2205.06548/) (2022)
- [PASS: Protected Attribute Suppression System for Mitigating Bias in Face Recognition](https://mlsanity.com/paper/2108.03764/) (2021)
- [Privacy-Preserving Face Recognition Using Trainable Feature Subtraction](https://mlsanity.com/paper/2403.12457/) (2024)
- [DuetFace: Collaborative Privacy-Preserving Face Recognition via Channel Splitting in the Frequency Domain](https://mlsanity.com/paper/2207.07340/) (2022)
- [Balancing Biases and Preserving Privacy on Balanced Faces in the Wild](https://mlsanity.com/paper/2103.09118/) (2021)
- [Privacy-Preserving Face Recognition Using Random Frequency Components](https://mlsanity.com/paper/2308.10461/) (2023)
- [Is Face Recognition Safe from Realizable Attacks?](https://mlsanity.com/paper/2210.08178/) (2022)
- [Intelligent Frame Selection as a Privacy-Friendlier Alternative to Face Recognition](https://mlsanity.com/paper/2101.07529/) (2021)
- [PrivacyProber: Assessment and Detection of Soft-Biometric Privacy-Enhancing Techniques](https://mlsanity.com/paper/2211.08864/) (2022)
- [A Critical Review of Predominant Bias in Neural Networks](https://mlsanity.com/paper/2502.11031/) (2025)

## Abstract (arXiv)

Recent research has proposed the use of Semi Adversarial Networks (SAN) for imparting privacy to face images. SANs are convolutional autoencoders that perturb face images such that the perturbed images cannot be reliably used by an attribute classifier (e.g., a gender classifier) but can still be used by a face matcher for matching purposes. However, the generalizability of SANs across multiple arbitrary gender classifiers has not been demonstrated in the literature. In this work, we tackle the generalization issue by designing an ensemble SAN model that generates a diverse set of perturbed outputs for a given input face image. This is accomplished by enforcing diversity among the individual models in the ensemble through the use of different data augmentation techniques. The goal is to ensure that at least one of the perturbed output faces will confound an arbitrary, previously unseen gender classifier. Extensive experiments using different unseen gender classifiers and face matchers are performed to demonstrate the efficacy of the proposed paradigm in imparting gender privacy to face images.