Back to Search Start Over

Uncertainty for Identifying Open-Set Errors in Visual Object Detection

Authors :
Miller, Dimity
Sünderhauf, Niko
Milford, Michael
Dayoub, Feras
Source :
IEEE Robotics and Automation Letters (January 2022), Volume 7, Issue 1, pages 215-222, ISSN 2377-3766
Publication Year :
2021

Abstract

Deployed into an open world, object detectors are prone to open-set errors, false positive detections of object classes not present in the training dataset. We propose GMM-Det, a real-time method for extracting epistemic uncertainty from object detectors to identify and reject open-set errors. GMM-Det trains the detector to produce a structured logit space that is modelled with class-specific Gaussian Mixture Models. At test time, open-set errors are identified by their low log-probability under all Gaussian Mixture Models. We test two common detector architectures, Faster R-CNN and RetinaNet, across three varied datasets spanning robotics and computer vision. Our results show that GMM-Det consistently outperforms existing uncertainty techniques for identifying and rejecting open-set detections, especially at the low-error-rate operating point required for safety-critical applications. GMM-Det maintains object detection performance, and introduces only minimal computational overhead. We also introduce a methodology for converting existing object detection datasets into specific open-set datasets to evaluate open-set performance in object detection.

Details

Database :
arXiv
Journal :
IEEE Robotics and Automation Letters (January 2022), Volume 7, Issue 1, pages 215-222, ISSN 2377-3766
Publication Type :
Report
Accession number :
edsarx.2104.01328
Document Type :
Working Paper
Full Text :
https://doi.org/10.1109/LRA.2021.3123374