Out-of-distribution Object Detection through Bayesian Uncertainty Estimation

Tianhao Zhang, Shenglin Wang, Nidhal Bouaynaya, Radu Calinescu, Lyudmila Mihaylova

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The superior performance of object detectors is often established under the condition that the test samples are in the same distribution as the training data. However, in many practical applications, out-of-distribution (OOD) instances are inevitable and usually lead to uncertainty in the results. In this paper, we propose a novel, intuitive, and scalable probabilistic object detection method for OOD detection. Unlike other uncertainty-modeling methods that either require huge computational costs to infer the weight distributions or rely on model training through synthetic outlier data, our method is able to distinguish between in-distribution (ID) data and OOD data via weight parameter sampling from proposed Gaussian distributions based on pre-trained networks. We demonstrate that our Bayesian object detector can achieve satisfactory OOD identification performance by reducing the FPR95 score by up to 8.19% and increasing the AUROC score by up to 13.94% when trained on BDD100k and VOC datasets as the ID datasets and evaluated on COCO2017 dataset as the OOD dataset.

Original languageEnglish (US)
Title of host publication2023 26th International Conference on Information Fusion, FUSION 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798890344854
DOIs
StatePublished - 2023
Externally publishedYes
Event26th International Conference on Information Fusion, FUSION 2023 - Charleston, United States
Duration: Jun 27 2023Jun 30 2023

Publication series

Name2023 26th International Conference on Information Fusion, FUSION 2023

Conference

Conference26th International Conference on Information Fusion, FUSION 2023
Country/TerritoryUnited States
CityCharleston
Period6/27/236/30/23

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Instrumentation

Fingerprint

Dive into the research topics of 'Out-of-distribution Object Detection through Bayesian Uncertainty Estimation'. Together they form a unique fingerprint.

Cite this