ImageNet

ImageNet is a large dataset of labelled images, organised according to the WordNet hierarchy of nouns. It was created by Fei-Fei Li and collaborators and first presented in 2009.

The full dataset has over 14 million images across more than 20,000 categories. A 1,000-class subset was used for the ImageNet Large Scale Visual Recognition Challenge (ILSVRC), which ran from 2010 to 2017 and became the standard benchmark for image classification.

AlexNet's win in the 2012 challenge, using a Convolutional Neural Network trained on GPUs, is often credited with kicking off the deep learning boom in computer vision.