无艺术背景生成模型：无需图形艺术知识的艺术创作

Art-Free Generative Models: Art Creation Without Graphic Art Knowledge

November 29, 2024

作者: Hui Ren, Joanna Materzynska, Rohit Gandikota, David Bau, Antonio Torralba

cs.AI

摘要

我们探讨一个问题：“创作艺术需要多少先前的艺术知识？”为了研究这个问题，我们提出了一个文本到图像生成模型，该模型在没有访问艺术相关内容的情况下进行训练。然后，我们引入了一种简单而有效的方法，仅使用少量选定艺术风格的示例来学习艺术适配器。我们的实验表明，使用我们的方法生成的艺术作品被用户认为与在大型、艺术丰富的数据集上训练的模型生成的艺术作品相媲美。最后，通过数据归因技术，我们阐明了来自艺术和非艺术数据集的示例如何促成了新艺术风格的创作。

English

We explore the question: "How much prior art knowledge is needed to create art?" To investigate this, we propose a text-to-image generation model trained without access to art-related content. We then introduce a simple yet effective method to learn an art adapter using only a few examples of selected artistic styles. Our experiments show that art generated using our method is perceived by users as comparable to art produced by models trained on large, art-rich datasets. Finally, through data attribution techniques, we illustrate how examples from both artistic and non-artistic datasets contributed to the creation of new artistic styles.