Learning to Generate Cross-Task Unexploitable Examples

TL;DR

This paper introduces a novel meta-learning framework for generating unexploitable images that are broadly resistant to various computer vision tasks, enhancing personal data privacy online.

Contribution

We propose the MCT-UEG framework with a flat-minima-oriented meta training scheme to improve the practicality of unexploitable example generation across multiple tasks.

Findings

The framework effectively generates broadly unexploitable images.

Extensive experiments validate the approach's efficacy.

The method enhances privacy protection in real-world scenarios.

Abstract

Unexploitable example generation aims to transform personal images into their unexploitable (unlearnable) versions before they are uploaded online, thereby preventing unauthorized exploitation of online personal images. Recently, this task has garnered significant research attention due to its critical relevance to personal data privacy. Yet, despite recent progress, existing methods for this task can still suffer from limited practical applicability, as they can fail to generate examples that are broadly unexploitable across different real-world computer vision tasks. To deal with this problem, in this work, we propose a novel Meta Cross-Task Unexploitable Example Generation (MCT-UEG) framework. At the core of our framework, to optimize the unexploitable example generator for effectively producing broadly unexploitable examples, we design a flat-minima-oriented meta training and testing scheme. Extensive experiments show the efficacy of our framework.