randall-lab/cub200

CUB-200-2011数据集是为了细粒度鸟类种类分类而设计的，包含了11,788张200种鸟类的图片，每种鸟类都有对应的标签。这个数据集在细粒度图像分类和计算机视觉研究中被广泛使用。每个样本包含一张可变尺寸的RGB图像和一个介于0到199之间的整数标签，代表鸟类的种类。数据集总共分为训练集和测试集，分别包含5,994张和5,794张图片。

The CUB-200-2011 dataset is designed for fine-grained bird species classification, containing 11,788 images of 200 bird species, each labeled with its corresponding category. This dataset is widely used in fine-grained image classification and computer vision research. Each sample consists of a variable-sized RGB image and an integer label between 0 and 199, representing the bird species. The dataset is split into a training set and a test set, with 5,994 and 5,794 images respectively.