OmniDepth: Dense Depth Estimation for Indoors Spherical Panoramas
This addresses the need for accurate depth estimation in 360-degree content, which is increasingly produced but lacks dedicated datasets, though it is incremental by repurposing existing data.
The paper tackled the problem of dense depth estimation for indoor spherical panoramas by creating a large-scale 360-degree dataset from existing 3D data and training a model, showing promising results on both synthesized and realistic images.
Recent work on depth estimation up to now has only focused on projective images ignoring 360 content which is now increasingly and more easily produced. We show that monocular depth estimation models trained on traditional images produce sub-optimal results on omnidirectional images, showcasing the need for training directly on 360 datasets, which however, are hard to acquire. In this work, we circumvent the challenges associated with acquiring high quality 360 datasets with ground truth depth annotations, by re-using recently released large scale 3D datasets and re-purposing them to 360 via rendering. This dataset, which is considerably larger than similar projective datasets, is publicly offered to the community to enable future research in this direction. We use this dataset to learn in an end-to-end fashion the task of depth estimation from 360 images. We show promising results in our synthesized data as well as in unseen realistic images.