I Find Your Lack of Uncertainty in Computer Vision Disturbing

TL;DR

This paper highlights the lack of proper epistemic uncertainty quantification in computer vision models, emphasizing its importance for high-stakes applications and urging the community to adopt better uncertainty estimation methods.

Contribution

It provides a meta-analysis showing most computer vision models ignore their limitations and advocates for integrating calibrated epistemic uncertainty estimation.

Findings

Most computer vision models lack proper uncertainty quantification

Ignoring uncertainty leads to risks in high-stakes applications

Recommendations for improving uncertainty estimation in the field

Abstract

Neural networks are used for many real world applications, but often they have problems estimating their own confidence. This is particularly problematic for computer vision applications aimed at making high stakes decisions with humans and their lives. In this paper we make a meta-analysis of the literature, showing that most if not all computer vision applications do not use proper epistemic uncertainty quantification, which means that these models ignore their own limitations. We describe the consequences of using models without proper uncertainty quantification, and motivate the community to adopt versions of the models they use that have proper calibrated epistemic uncertainty, in order to enable out of distribution detection. We close the paper with a summary of challenges on estimating uncertainty for computer vision applications and recommendations.