The dataset used in the paper is not explicitly described, but it is mentioned that the authors used CIFAR-10, CIFAR-100, and STL-10 datasets for training and testing the embedding functions.
BibTex:
Before browse our site, please accept our cookies policy