Facebook has just introduced a great dataset and corresponding machine learning challenge at same time. The dataset is named LVIS (Large Vocabulary Instance Segmentation).
This is a great visual instance dataset.
The challenge: https://www.lvisdataset.org/challenge
Today, rigorous evaluation of general purpose object detectors is mostly performed in the few category regime (e.g. 80) or when there are a large number of training examples per category (e.g. 100 to 1000+). LVIS provides an opportunity to enable research in the setting where there are a large number of categories and where per-category data is sometimes scarce.
Given that state-of-the-art deep learning methods for object detection perform poorly in the low-sample regime, we believe that our dataset poses an important and exciting new scientific challenge.
The dataset: https://www.lvisdataset.org/dataset
Source: https://twitter.com/facebookai/status/1159548405867139074