-
Flowers102
The dataset used in the paper is Flowers102, which is a natural image classification dataset. It contains 102 categories of natural images. -
Caltech101
The dataset used in the paper is Caltech101, which is a natural image classification dataset. It contains 101 categories of natural images.