No Way To Steal My Face: Proactive Defense Against Identity-Preserving Personalized Generation

Summary

IDGuardian is a model-agnostic identity protection framework that defends against both training-based and training-free personalized image generation with diffusion models. It splits personalization into identity extraction and identity injection, then adds adversarial perturbations to disrupt both stages. The authors report that it protects identity across various pipelines and architectures and stays robust to post-processing, adaptive attacks, and cross-dataset generalization.