# Context The data is based on images I have taken with my Lytro Illum camera (https://pictures.lytro.com/ksmader) they have been exported as image data and depth maps. The idea is to make and build tools for looking at Lytro Image data and improving the results # Content The data are from the Lytro Illum and captured as 40MP images which are then converted to 5MP RGB+D images. All of the required data for several test images is provided The second datasets come from the Lenovo Phab2 (Project Tango) which utilizes dual image sensors to recreate point clouds of large 3D structures. These are provided as .ply and .obj datasets # Acknowledgements The data is based on images I have taken with my Lytro Illum camera (https://pictures.lytro.com/ksmader). # Inspiration 1. Build a neural network which automatically generates depth information from 2D RGB images 2. Build a tool to find gaps or holes in the depth images and fixes them automatically 3. Build a neural network which can reconstruct 3D pixel data from RGBD images