DFA-CON: A Contrastive Learning Approach for Detecting Copyright Infringement in DeepFake Art

TL;DR

DFA-CON is a contrastive learning framework that effectively detects copyright infringement and forgery in AI-generated art, addressing challenges posed by various attack types and outperforming existing models.

Contribution

This work introduces DFA-CON, a novel contrastive learning approach specifically designed for detecting copyright violations in AI-generated visual artworks.

Findings

Robust detection across multiple attack types

Outperforms recent pretrained foundation models

Effective discrimination between original and forged artworks

Abstract

Recent proliferation of generative AI tools for visual content creation-particularly in the context of visual artworks-has raised serious concerns about copyright infringement and forgery. The large-scale datasets used to train these models often contain a mixture of copyrighted and non-copyrighted artworks. Given the tendency of generative models to memorize training patterns, they are susceptible to varying degrees of copyright violation. Building on the recently proposed DeepfakeArt Challenge benchmark, this work introduces DFA-CON, a contrastive learning framework designed to detect copyright-infringing or forged AI-generated art. DFA-CON learns a discriminative representation space, posing affinity among original artworks and their forged counterparts within a contrastive learning framework. The model is trained across multiple attack types, including inpainting, style transfer, adversarial perturbation, and cutmix. Evaluation results demonstrate robust detection performance across most attack types, outperforming recent pretrained foundation models. Code and model checkpoints will be released publicly upon acceptance.