Uncertainty Estimation in One-Stage Object Detection

TL;DR

This paper introduces a method for estimating uncertainty in one-stage object detection models, enhancing detection accuracy and providing valuable confidence measures for autonomous vehicle perception tasks.

Contribution

It proposes a novel approach to quantify uncertainty in one-stage detectors, improving detection performance and enabling better decision-making in autonomous driving.

Findings

Uncertainty estimates correlate with detection accuracy.

The approach improves baseline detection performance.

Uncertainty relates to pedestrian occlusion levels.

Abstract

Environment perception is the task for intelligent vehicles on which all subsequent steps rely. A key part of perception is to safely detect other road users such as vehicles, pedestrians, and cyclists. With modern deep learning techniques huge progress was made over the last years in this field. However such deep learning based object detection models cannot predict how certain they are in their predictions, potentially hampering the performance of later steps such as tracking or sensor fusion. We present a viable approaches to estimate uncertainty in an one-stage object detector, while improving the detection performance of the baseline approach. The proposed model is evaluated on a large scale automotive pedestrian dataset. Experimental results show that the uncertainty outputted by our system is coupled with detection accuracy and the occlusion level of pedestrians.

Full text

Uncertainty Estimation in One-Stage Object Detection

Florian Kraus1 and Klaus Dietmayer2 1Florian Kraus is with Daimler AG, 89081 Ulm, Germany, 2Klaus Dietmayer is with Institute of Measurement, Control and Microtechnology, Ulm University, 89081 Ulm, Germany

abstract introduction related_work method experiments conclusion_and_acknowledgments