DRO: Deep Recurrent Optimizer for Video to Depth

There are increasing interests of studying the video-to-depth (V2D) problem with machine learning techniques. While earlier methods directly learn a mapping from images to depth maps and camera poses, more recent works enforce multi-view geometry constraints through optimization embedded in framework. This paper presents novel method based on recurrent neural networks further exploit potential V2D. Specifically, our optimizer alternately updates poses iterations minimize feature-metric cost, two gated units iteratively improve results by tracing historical information. Extensive experimental demonstrate that outperforms previous is efficient computation memory consumption than cost-volume-based methods. In particular, self-supervised supervised KITTI ScanNet datasets. Our source code will be made public.