Visual-Friendly Concept Protection via Selective Adversarial Perturbations

TL;DR

This paper introduces VCPro, a framework that creates less perceptible adversarial perturbations to protect specific concepts in images, balancing protection effectiveness with visual quality.

Contribution

It proposes a novel relaxed optimization approach to generate inconspicuous adversarial perturbations for concept protection in images.

Findings

VCPro achieves a better trade-off between perturbation visibility and protection effectiveness.

The method effectively prioritizes target concept protection with minimal perceptibility.

Experiments demonstrate improved visual quality while maintaining protection performance.

Abstract

Personalized concept generation by tuning diffusion models with a few images raises potential legal and ethical concerns regarding privacy and intellectual property rights. Researchers attempt to prevent malicious personalization using adversarial perturbations. However, previous efforts have mainly focused on the effectiveness of protection while neglecting the visibility of perturbations. They utilize global adversarial perturbations, which introduce noticeable alterations to original images and significantly degrade visual quality. In this work, we propose the Visual-Friendly Concept Protection (VCPro) framework, which prioritizes the protection of key concepts chosen by the image owner through adversarial perturbations with lower perceptibility. To ensure these perturbations are as inconspicuous as possible, we introduce a relaxed optimization objective to identify the least perceptible yet effective adversarial perturbations, solved using the Lagrangian multiplier method. Qualitative and quantitative experiments validate that VCPro achieves a better trade-off between the visibility of perturbations and protection effectiveness, effectively prioritizing the protection of target concepts in images with less perceptible perturbations.