Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Assessing Box Merging Strategies and Uncertainty Estimation Methods in Multimodel Object Detection

: Schmoeller Roza, Felippe; Henne, Maximilian; Roscher, Karsten; Günnemann, Stephan

Postprint urn:nbn:de:0011-n-6214634 (567 KByte PDF)
MD5 Fingerprint: d3949ac1992cb0e4395e8edc6f09d313
The original publication is available at
Erstellt am: 10.2.2021

Bartoli, Adrien (Ed.):
Computer Vision - ECCV 2020 Workshops. Proceedings. Pt.VI : Glasgow, UK, August 23-28, 2020
Cham: Springer Nature, 2020 (Lecture Notes in Computer Science 12540)
ISBN: 978-3-030-65413-9 (Print)
ISBN: 978-3-030-65414-6 (Online)
European Conference on Computer Vision (ECCV) <16, 2020, Online>
Workshop "Beyond mAP - Reassessing the Evaluation of Object Detectors" <2020, Online>
Bayerisches Staatsministerium für Wirtschaft, Landesentwicklung und Energie StMWi

Konferenzbeitrag, Elektronische Publikation
Fraunhofer IKS ()
uncertainty estimation; deep ensembles; object detection

This paper examines the impact of different box merging strategies for sampling-based uncertainty estimation methods in object detection. Also, a comparison between the almost exclusively used softmax confidence scores and the predicted variances on the quality of the final predictions estimates is presented. The results suggest that estimated variances are a stronger predictor for the detection quality. However, variance-based merging strategies do not improve significantly over the confidence-based alternative for the given setup. In contrast, we show that different methods to estimate the uncertainty of the predictions have a significant influence on the quality of the ensembling outcome. Since mAP does not reward uncertainty estimates, such improvements were only noticeable on the resulting PDQ scores.