SUN Dataset is a dataset for scene understanding, which covers 130,519 images in 899 categories. The publisher uses 397 well-sampled categories for scene recognition and uses them with state-of-the-art algorithms to establish new performance boundaries.

This dataset was released by the Princeton Vision and Robotics Laboratory in 2014. Related papers include "SUN Database: Large-scale Scene Recognition from Abbey to Zoo" and "SUN Database: Exploring a Large Collection of Scene Categories".