This dataset was collected with a single hand-held Lytro Illum camera during my 2017 internship with the Stanford Computational Imaging. The dataset contains 4211 light fields organized into 30 categories (i.e. trees, benches), and each scene is captured from 4-6 camera poses.