Mitsua/art-museums-pd-440k

这是一个用于训练文本到图像或任何文本和图像多模态模型的数据集，旨在最小化版权/许可问题。数据集中的所有图像和文本均在CC0或公共领域下共享，且未使用任何预训练模型或AI模型来构建此数据集，除了用于将英文标题翻译成日文的ElanMT模型。图像和元数据来自多个博物馆的开放访问资源，包括史密森尼开放访问、大都会艺术博物馆开放访问、克利夫兰艺术博物馆开放访问和芝加哥艺术学院开放访问。数据集还进行了基于单词的预过滤，以减轻有害或不适当内容的影响。图像和原始元数据由这些博物馆在CC0下许可，而数据集本身和编译/翻译的标题则由ELAN MITSUA项目/Abstract Engine在CC BY 4.0下许可。

This is a dataset to train text-to-image or any text and image multimodal models with minimized copyright/licensing concerns. All images and texts in this dataset are originally shared under CC0 or public domain, and no pretrained models or any AI models are used to build this dataset except for our ElanMT model to translate English captions to Japanese. The data sources include open access data from multiple museums, with all images and metadata shared under CC0 or Public Domain. The dataset itself and compiled/translated captions are licensed under CC BY 4.0.