A Theoretical and Practical Framework for Evaluating Uncertainty Calibration in Object Detection

TL;DR

This paper introduces a comprehensive theoretical and practical framework for evaluating uncertainty calibration in object detection systems, crucial for safety-critical applications like autonomous driving and medical diagnosis.

Contribution

It provides a new formal foundation and three novel metrics for assessing uncertainty calibration in object detection, supported by experimental validation.

Findings

Proposed a formal definition of uncertainty calibration for object detection.

Developed three new evaluation metrics for calibration assessment.

Validated metrics through representative experiments.

Abstract

The proliferation of Deep Neural Networks has resulted in machine learning systems becoming increasingly more present in various real-world applications. Consequently, there is a growing demand for highly reliable models in many domains, making the problem of uncertainty calibration pivotal when considering the future of deep learning. This is especially true when considering object detection systems, that are commonly present in safety-critical applications such as autonomous driving, robotics and medical diagnosis. For this reason, this work presents a novel theoretical and practical framework to evaluate object detection systems in the context of uncertainty calibration. This encompasses a new comprehensive formulation of this concept through distinct formal definitions, and also three novel evaluation metrics derived from such theoretical foundation. The robustness of the proposed uncertainty calibration metrics is shown through a series of representative experiments.