Dayton
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/kregmi/cross-view-image-synthesis/blob/master/readme.md
下载链接
链接失效反馈官方服务：
资源简介：
该数据集名为Dayton，包含了为跨视角图像翻译任务训练和测试而处理的图像对。所有图像均已调整为256x256的分辨率。该数据集规模为76,048张图像，分为训练集和测试集，其中训练集有55,000对，测试集有21,048对，所涉及的任务是跨视角图像翻译。
The dataset named Dayton consists of image pairs processed for training and evaluating cross-view image translation tasks. All images have been resized to a resolution of 256×256. The dataset contains a total of 76,048 images, split into training and test subsets: the training subset holds 55,000 image pairs, and the test subset includes 21,048 image pairs. The core task supported by this dataset is cross-view image translation.
