EditShield: Protecting Unauthorized Image Editing by Instruction-guided Diffusion Models

TL;DR

This paper introduces EditShield, a method that adds imperceptible perturbations to images to prevent unauthorized editing by instruction-guided diffusion models, ensuring image integrity.

Contribution

The paper presents EditShield, a novel protection technique against unauthorized image modifications by instruction-guided diffusion models, by shifting latent representations to produce unrealistic outputs.

Findings

EditShield effectively prevents unauthorized edits in synthetic and real-world datasets.

It remains robust across different editing types and instruction phrases.

The method successfully maintains imperceptibility of perturbations.

Abstract

Text-to-image diffusion models have emerged as an evolutionary for producing creative content in image synthesis. Based on the impressive generation abilities of these models, instruction-guided diffusion models can edit images with simple instructions and input images. While they empower users to obtain their desired edited images with ease, they have raised concerns about unauthorized image manipulation. Prior research has delved into the unauthorized use of personalized diffusion models; however, this problem of instruction-guided diffusion models remains largely unexplored. In this paper, we first propose a protection method EditShield against unauthorized modifications from such models. Specifically, EditShield works by adding imperceptible perturbations that can shift the latent representation used in the diffusion process, tricking models into generating unrealistic images with mismatched subjects. Our extensive experiments demonstrate EditShield's effectiveness among synthetic and real-world datasets. Besides, we found that EditShield performs robustly against various manipulation settings across editing types and synonymous instruction phrases.