A New Distributional Ranking Loss With Uncertainty: Illustrated in Relative Depth Estimation

TL;DR

This paper introduces a probabilistic approach to relative depth estimation from a single image, using a novel distributional ranking loss that provides confidence measures and achieves state-of-the-art results.

Contribution

It presents a new distributional ranking loss and a probabilistic model for depth estimation that outputs confidence levels, improving accuracy and reliability.

Findings

Achieved state-of-the-art results on depth estimation benchmarks.

Model's confidence correlates well with estimation accuracy.

Using confidence improves downstream depth estimation tasks.

Abstract

We propose a new approach for the problem of relative depth estimation from a single image. Instead of directly regressing over depth scores, we formulate the problem as estimation of a probability distribution over depth and aim to learn the parameters of the distributions which maximize the likelihood of the given data. To train our model, we propose a new ranking loss, Distributional Loss, which tries to increase the probability of farther pixel's depth being greater than the closer pixel's depth. Our proposed approach allows our model to output confidence in its estimation in the form of standard deviation of the distribution. We achieve state of the art results against a number of baselines while providing confidence in our estimations. Our analysis show that estimated confidence is actually a good indicator of accuracy. We investigate the usage of confidence information in a downstream task of metric depth estimation, to increase its performance.