Relative Depth Estimation as a Ranking Problem
This work addresses depth estimation for computer vision applications, but it is incremental as it adapts existing ranking methods to a known task.
The authors tackled the problem of relative depth estimation from a single image by reformulating it as a ranking problem, achieving improved results through the application of a listwise ranking loss and a new metric for pixel depth ranking accuracy.
We present a formulation of the relative depth estimation from a single image problem, as a ranking problem. By reformulating the problem this way, we were able to utilize literature on the ranking problem, and apply the existing knowledge to achieve better results. To this end, we have introduced a listwise ranking loss borrowed from ranking literature, weighted ListMLE, to the relative depth estimation problem. We have also brought a new metric which considers pixel depth ranking accuracy, on which our method is stronger.