Dress Code

该数据集名为“Dress Code”，是一个高分辨率图像为基础的虚拟试衣数据集，它包含了多类别衣物与全身参考模型的配对图片，旨在解决现有数据集的局限性。该数据集涵盖了上装、下装和连衣裙三个类别的图片，所有图片均为高分辨率（1024x768像素），并包含了一个自动选择有效图片对的过程。该数据集规模宏大，拥有超过10万张图片，其中包括53,795对配对图片，其任务是实现基于图像的虚拟试衣。

This dataset, named *Dress Code*, is a high-resolution image-based virtual try-on dataset. It contains paired images of multi-category clothing items and full-body reference models, aiming to address the limitations of existing datasets. Covering three clothing categories: tops, bottoms, and dresses, all images in this dataset have a resolution of 1024×768 pixels and incorporate an automated process for selecting valid image pairs. With a large-scale corpus, the dataset includes over 100,000 images in total, among which there are 53,795 paired image samples. Its core task is to enable image-based virtual try-on.