Protect-Your-IP: Scalable Source-Tracing and Attribution against Personalized Generation

TL;DR

This paper introduces a scalable, unified watermarking and attribution framework for AI-generated images, effectively tracing sources and identifying personalized models while maintaining robustness and adaptability.

Contribution

It presents an innovative combined proactive and passive watermarking approach with incremental learning for source-tracing and attribution of personalized AI-generated images.

Findings

Effective source-tracing and attribution demonstrated on celebrity portrait datasets.

Robustness against knowledge forgetting in incremental learning.

High accuracy in distinguishing AI-generated images from authentic ones.

Abstract

With the advent of personalized generation models, users can more readily create images resembling existing content, heightening the risk of violating portrait rights and intellectual property (IP). Traditional post-hoc detection and source-tracing methods for AI-generated content (AIGC) employ proactive watermark approaches; however, these are less effective against personalized generation models. Moreover, attribution techniques for AIGC rely on passive detection but often struggle to differentiate AIGC from authentic images, presenting a substantial challenge. Integrating these two processes into a cohesive framework not only meets the practical demands for protection and forensics but also improves the effectiveness of attribution tasks. Inspired by this insight, we propose a unified approach for image copyright source-tracing and attribution, introducing an innovative watermarking-attribution method that blends proactive and passive strategies. We embed copyright watermarks into protected images and train a watermark decoder to retrieve copyright information from the outputs of personalized models, using this watermark as an initial step for confirming if an image is AIGC-generated. To pinpoint specific generation techniques, we utilize powerful visual backbone networks for classification. Additionally, we implement an incremental learning strategy to adeptly attribute new personalized models without losing prior knowledge, thereby enhancing the model's adaptability to novel generation methods. We have conducted experiments using various celebrity portrait series sourced online, and the results affirm the efficacy of our method in source-tracing and attribution tasks, as well as its robustness against knowledge forgetting.