Stable Diffusion Wordnet Dataset

Stable Diffusion Wordnet Dataset是由上奥地利应用科技大学Andreas Stöckl创建的一个合成图像数据集，包含262,040张由Stable Diffusion模型生成的图像。该数据集利用Wordnet分类法和概念定义作为生成图像的文本提示，旨在用于机器学习应用中的数据增强。数据集的创建过程涉及使用Wordnet中的名词synsets作为文本输入，每synset生成10张图像。该数据集适用于评估Stable Diffusion模型的能力，特别是在图像分类任务中，以及用于探索合成数据在监督学习设置中的多样性增强。

The Stable Diffusion Wordnet Dataset is a synthetic image dataset created by Andreas Stöckl from the University of Applied Sciences Upper Austria, consisting of 262,040 images generated by the Stable Diffusion model. This dataset uses Wordnet taxonomy and concept definitions as text prompts for image generation, and is intended for data augmentation in machine learning applications. The dataset creation process employs noun synsets from Wordnet as text inputs, generating 10 images per synset. This dataset is suitable for evaluating the capabilities of the Stable Diffusion model, particularly in image classification tasks, as well as exploring the diversity enhancement of synthetic data in supervised learning settings.