Measuring and Modeling Uncertainty Degree for Monocular Depth Estimation

TL;DR

This paper introduces a simple yet effective approach to measure and model uncertainty in monocular depth estimation by leveraging inherent probability distributions, achieving state-of-the-art reliability without extra modules.

Contribution

It proposes a novel uncertainty modeling method based on depth probability distributions, improving reliability with minimal additional training complexity.

Findings

Achieves state-of-the-art uncertainty reliability in MDE

No extra modules or multiple inferences needed

Enhances performance with ensemble or sampling methods

Abstract

Effectively measuring and modeling the reliability of a trained model is essential to the real-world deployment of monocular depth estimation (MDE) models. However, the intrinsic ill-posedness and ordinal-sensitive nature of MDE pose major challenges to the estimation of uncertainty degree of the trained models. On the one hand, utilizing current uncertainty modeling methods may increase memory consumption and are usually time-consuming. On the other hand, measuring the uncertainty based on model accuracy can also be problematic, where uncertainty reliability and prediction accuracy are not well decoupled. In this paper, we propose to model the uncertainty of MDE models from the perspective of the inherent probability distributions originating from the depth probability volume and its extensions, and to assess it more fairly with more comprehensive metrics. By simply introducing additional training regularization terms, our model, with surprisingly simple formations and without requiring extra modules or multiple inferences, can provide uncertainty estimations with state-of-the-art reliability, and can be further improved when combined with ensemble or sampling methods. A series of experiments demonstrate the effectiveness of our methods.