Robust Semi-Supervised Monocular Depth Estimation with Reprojected Distances
This work addresses the challenge of scale-aware depth estimation for robotics applications, offering an incremental improvement by reducing the need for dense supervision.
The paper tackles the problem of monocular depth estimation by combining supervised and self-supervised approaches, proposing a novel supervised loss term that enables training with very sparse LiDAR data (as few as 4 beams, less than 100 depth values per image) to achieve results competitive with state-of-the-art.
Dense depth estimation from a single image is a key problem in computer vision, with exciting applications in a multitude of robotic tasks. Initially viewed as a direct regression problem, requiring annotated labels as supervision at training time, in the past few years a substantial amount of work has been done in self-supervised depth training based on strong geometric cues, both from stereo cameras and more recently from monocular video sequences. In this paper we investigate how these two approaches (supervised & self-supervised) can be effectively combined, so that a depth model can learn to encode true scale from sparse supervision while achieving high fidelity local accuracy by leveraging geometric cues. To this end, we propose a novel supervised loss term that complements the widely used photometric loss, and show how it can be used to train robust semi-supervised monocular depth estimation models. Furthermore, we evaluate how much supervision is actually necessary to train accurate scale-aware monocular depth models, showing that with our proposed framework, very sparse LiDAR information, with as few as 4 beams (less than 100 valid depth values per image), is enough to achieve results competitive with the current state-of-the-art.