IBims-1

该数据集包含了100对高质量的深度图和高分辨率图像。值得注意的是，该数据集没有训练集，而是通过在NYUDv2数据集上训练的模型来进行泛化能力评估。规模上，数据集包含100对图像，任务专注于单目深度估计。

This dataset contains 100 high-quality pairs of depth maps and high-resolution images. Notably, this dataset does not include a training set, and is instead used to evaluate the generalization capability of models trained on the NYUDv2 dataset. In terms of scale, the dataset consists of 100 image pairs, and the task focuses on monocular depth estimation.