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.
Abstract (translated)
近期，用于视觉内容创作的生成式人工智能工具（特别是针对视觉艺术作品）的普及引发了关于版权侵权和伪造的严重担忧。训练这些模型所使用的大型数据集通常包含受版权保护的艺术作品与不受版权保护的作品混合在一起的情况。鉴于生成性模型倾向于记住训练模式，它们在不同程度上容易发生版权侵犯行为。 在此背景下，本工作借鉴了最近提出的DeepfakeArt挑战基准测试，并引入了一种名为DFA-CON的对比学习框架，旨在检测侵权或伪造的人工智能生成的艺术作品。该框架通过构建一个区分性的表示空间，在对比学习框架内促进原始艺术作品与其被伪造版本之间的关联性。 模型在多种攻击类型下进行训练，包括图像修复（inpainting）、风格转换、对抗性扰动和混合剪切（cutmix）。评估结果显示，在大多数攻击类型下，该方法具有强大的检测性能，并且优于最近预训练的基础模型。本研究的代码和模型检查点将在接受后公开发布。
URL
https://arxiv.org/abs/2505.08552
