Toward Privacy and Utility Preserving Image Representation

TL;DR

This paper introduces the Adversarial Image Anonymizer (AIA), a framework that creates privacy-preserving image representations by balancing privacy protection with utility for face verification tasks.

Contribution

The paper proposes a novel adversarial learning framework, AIA, for generating image representations that simultaneously preserve privacy and utility, addressing a less-explored aspect of privacy in face images.

Findings

AIA effectively balances privacy and utility in face image representations.

Experimental results demonstrate AIA's superiority over existing methods.

AIA maintains high utility for face verification while enhancing privacy protection.

Abstract

Face images are rich data items that are useful and can easily be collected in many applications, such as in 1-to-1 face verification tasks in the domain of security and surveillance systems. Multiple methods have been proposed to protect an individual's privacy by perturbing the images to remove traces of identifiable information, such as gender or race. However, significantly less attention has been given to the problem of protecting images while maintaining optimal task utility. In this paper, we study the novel problem of creating privacy-preserving image representations with respect to a given utility task by proposing a principled framework called the Adversarial Image Anonymizer (AIA). AIA first creates an image representation using a generative model, then enhances the learned image representations using adversarial learning to preserve privacy and utility for a given task. Experiments were conducted on a publicly available data set to demonstrate the effectiveness of AIA as a privacy-preserving mechanism for face images.