Anonymizing k-Facial Attributes via Adversarial Perturbations

TL;DR

This paper introduces an adversarial perturbation method that selectively anonymizes facial attributes like gender and age in images, protecting privacy while maintaining image quality and identity.

Contribution

It presents a novel algorithm that allows users to anonymize specific facial attributes without affecting image identity or visual quality.

Findings

Successfully anonymizes multiple attributes in face images

Preserves image quality and identity information

Effective on multiple popular datasets

Abstract

A face image not only provides details about the identity of a subject but also reveals several attributes such as gender, race, sexual orientation, and age. Advancements in machine learning algorithms and popularity of sharing images on the World Wide Web, including social media websites, have increased the scope of data analytics and information profiling from photo collections. This poses a serious privacy threat for individuals who do not want to be profiled. This research presents a novel algorithm for anonymizing selective attributes which an individual does not want to share without affecting the visual quality of images. Using the proposed algorithm, a user can select single or multiple attributes to be surpassed while preserving identity information and visual content. The proposed adversarial perturbation based algorithm embeds imperceptible noise in an image such that attribute prediction algorithm for the selected attribute yields incorrect classification result, thereby preserving the information according to user's choice. Experiments on three popular databases i.e. MUCT, LFWcrop, and CelebA show that the proposed algorithm not only anonymizes k-attributes, but also preserves image quality and identity information.