LSUN Dataset is a large-scale image dataset built on top of deep learning by humans in the loop, containing 10 scene categories and 20 object categories, totaling about 1 million labeled images.

This dataset was released by the University of California, Berkeley in 2015. The main publishers are Fisher Yu, Peter Kontschieder, Shuran Song, Ming Jiang, Yinda Zhang, Catherine Qi Zhao, Thomas Funkhouser and Jianxiong Xiao. The related paper is "LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop".

The dataset’s related challenges focus on scene understanding, including scene object retrieval, outdoor scene segmentation, RGB-D 3D object detection, and saliency detection.