copyright Plug-in Authorization for Human Content Copyright Protection in Text-to-Image Model

2024-04-18

arXiv.org

AI Summary

To address copyright infringement in text-to-image generation, this paper proposes ©-Plugin, a fine-grained, traceable, revocable, and composable copyright authorization framework. Methodologically, it introduces Reverse LoRA—a novel technique for reverse-engineering stylistic or intellectual property (IP) features from pre-trained models—and integrates it with EasyMerge, a low-rank fusion method enabling seamless embedding of copyright modules into both training and inference pipelines. The framework unifies LoRA-based fine-tuning, reverse gradient optimization, and style/IP-directed distillation to support cross-model copyright transfer and precise attribution. Extensive evaluation on artist-style replication and cartoon IP reconstruction demonstrates its effectiveness. The implementation is publicly available.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Measurement, analysis, and circumvention of Web censorship
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

This paper addresses the contentious issue of copyright infringement in images generated by text-to-image models, sparking debates among AI developers, content creators, and legal entities. State-of-the-art models create high-quality content without crediting original creators, causing concern in the artistic community. To mitigate this, we propose the copyright Plug-in Authorization framework, introducing three operations: addition, extraction, and combination. Addition involves training a copyright plug-in for specific copyright, facilitating proper credit attribution. Extraction allows creators to reclaim copyright from infringing models, and combination enables users to merge different copyright plug-ins. These operations act as permits, incentivizing fair use and providing flexibility in authorization. We present innovative approaches,"Reverse LoRA"for extraction and"EasyMerge"for seamless combination. Experiments in artist-style replication and cartoon IP recreation demonstrate copyright plug-ins' effectiveness, offering a valuable solution for human copyright protection in the age of generative AIs. The code is available at https://github.com/zc1023/-Plug-in-Authorization.git.

Problem

• Copyright Protection
• Text-to-Image Models
• Artists' Rights

Innovation

• Copyright Plugin Framework
• Reverse LoRA
• EasyMerge