VITON

该数据集名为VITON，包含了一组女性模特穿着不同上衣的前视图图像，共计16,253张，并分为训练集和测试集两部分。该数据集主要用于训练生成对抗网络（GAN）模型，并且包含一个独立的测试集，以便进行性能评估。整个数据集的规模为16,253张图像，其任务领域为时尚图像编辑。

This dataset, named VITON, contains 16,253 front-view images of female models wearing various upper garments, and is split into training and test sets. It is primarily utilized for training Generative Adversarial Network (GAN) models, and features an independent test subset dedicated to model performance evaluation. With a total of 16,253 images, this dataset targets the task domain of fashion image editing.