360D: A dataset and baseline for dense depth estimation from 360 images
We present a baseline for 360 dense depth estimation from a single spherical panorama. We circumvent the unavailability of coupled 360 color and depth image datasets by rendering a high quality 360 dataset from existing 3D datasets. We then train a CNN designed specifically for 360 content in a supervised manner, in order to predict a 360o depth map from a single omnidirectional image in equirectangular format. Quantitative and qualitative results show the need for training directly in 360 instead of relying on traditional 2D CNNs.