Nearly Zero-Cost Protection Against Mimicry by Personalized Diffusion Models

TL;DR

This paper presents a new method for protecting images generated by diffusion models from mimicry and misuse, achieving near-zero-cost protection with high invisibility and low latency.

Contribution

It introduces perturbation pre-training and a mixture-of-perturbations approach for adaptive, efficient image protection in diffusion models.

Findings

Protection performance is comparable to existing methods.

Invisibility of protected images is significantly improved.

Inference time is drastically reduced.

Abstract

Recent advancements in diffusion models revolutionize image generation but pose risks of misuse, such as replicating artworks or generating deepfakes. Existing image protection methods, though effective, struggle to balance protection efficacy, invisibility, and latency, thus limiting practical use. We introduce perturbation pre-training to reduce latency and propose a mixture-of-perturbations approach that dynamically adapts to input images to minimize performance degradation. Our novel training strategy computes protection loss across multiple VAE feature spaces, while adaptive targeted protection at inference enhances robustness and invisibility. Experiments show comparable protection performance with improved invisibility and drastically reduced inference time. The code and demo are available at https://webtoon.github.io/impasto