Cost-Sensitive Uncertainty-Based Failure Recognition for Object Detection

Moussa Kassem-Sbeyti, Michelle Karg, Christian Wirth, Nadja Klein, Sahin Albayrak
Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, PMLR 244:1890-1900, 2024.

Abstract

Object detectors in real-world applications often fail to detect objects due to varying factors such as weather conditions and noisy input. Therefore, a process that mitigates false detections is crucial for both safety and accuracy. While uncertainty-based thresholding shows promise, previous works demonstrate an imperfect correlation between uncertainty and detection errors. This hinders ideal thresholding, prompting us to further investigate the correlation and associated cost with different types of uncertainty. We therefore propose a cost-sensitive framework for object detection tailored to user-defined budgets on the two types of errors, missing and false detections. We derive minimum thresholding requirements to prevent performance degradation and define metrics to assess the applicability of uncertainty for failure recognition. Furthermore, we automate and optimize the thresholding process to maximize the failure recognition rate w.r.t. the specified budget. Evaluation on three autonomous driving datasets demonstrates that our approach significantly enhances safety, particularly in challenging scenarios. Leveraging localization aleatoric uncertainty and softmax-based entropy only, our method boosts the failure recognition rate by 36-60% compared to conventional approaches. Code is available at https://mos-ks.github.io/publications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v244-kassem-sbeyti24a, title = {Cost-Sensitive Uncertainty-Based Failure Recognition for Object Detection}, author = {Kassem-Sbeyti, Moussa and Karg, Michelle and Wirth, Christian and Klein, Nadja and Albayrak, Sahin}, booktitle = {Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence}, pages = {1890--1900}, year = {2024}, editor = {Kiyavash, Negar and Mooij, Joris M.}, volume = {244}, series = {Proceedings of Machine Learning Research}, month = {15--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v244/main/assets/kassem-sbeyti24a/kassem-sbeyti24a.pdf}, url = {https://proceedings.mlr.press/v244/kassem-sbeyti24a.html}, abstract = {Object detectors in real-world applications often fail to detect objects due to varying factors such as weather conditions and noisy input. Therefore, a process that mitigates false detections is crucial for both safety and accuracy. While uncertainty-based thresholding shows promise, previous works demonstrate an imperfect correlation between uncertainty and detection errors. This hinders ideal thresholding, prompting us to further investigate the correlation and associated cost with different types of uncertainty. We therefore propose a cost-sensitive framework for object detection tailored to user-defined budgets on the two types of errors, missing and false detections. We derive minimum thresholding requirements to prevent performance degradation and define metrics to assess the applicability of uncertainty for failure recognition. Furthermore, we automate and optimize the thresholding process to maximize the failure recognition rate w.r.t. the specified budget. Evaluation on three autonomous driving datasets demonstrates that our approach significantly enhances safety, particularly in challenging scenarios. Leveraging localization aleatoric uncertainty and softmax-based entropy only, our method boosts the failure recognition rate by 36-60% compared to conventional approaches. Code is available at https://mos-ks.github.io/publications.} }
Endnote
%0 Conference Paper %T Cost-Sensitive Uncertainty-Based Failure Recognition for Object Detection %A Moussa Kassem-Sbeyti %A Michelle Karg %A Christian Wirth %A Nadja Klein %A Sahin Albayrak %B Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2024 %E Negar Kiyavash %E Joris M. Mooij %F pmlr-v244-kassem-sbeyti24a %I PMLR %P 1890--1900 %U https://proceedings.mlr.press/v244/kassem-sbeyti24a.html %V 244 %X Object detectors in real-world applications often fail to detect objects due to varying factors such as weather conditions and noisy input. Therefore, a process that mitigates false detections is crucial for both safety and accuracy. While uncertainty-based thresholding shows promise, previous works demonstrate an imperfect correlation between uncertainty and detection errors. This hinders ideal thresholding, prompting us to further investigate the correlation and associated cost with different types of uncertainty. We therefore propose a cost-sensitive framework for object detection tailored to user-defined budgets on the two types of errors, missing and false detections. We derive minimum thresholding requirements to prevent performance degradation and define metrics to assess the applicability of uncertainty for failure recognition. Furthermore, we automate and optimize the thresholding process to maximize the failure recognition rate w.r.t. the specified budget. Evaluation on three autonomous driving datasets demonstrates that our approach significantly enhances safety, particularly in challenging scenarios. Leveraging localization aleatoric uncertainty and softmax-based entropy only, our method boosts the failure recognition rate by 36-60% compared to conventional approaches. Code is available at https://mos-ks.github.io/publications.
APA
Kassem-Sbeyti, M., Karg, M., Wirth, C., Klein, N. & Albayrak, S.. (2024). Cost-Sensitive Uncertainty-Based Failure Recognition for Object Detection. Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 244:1890-1900 Available from https://proceedings.mlr.press/v244/kassem-sbeyti24a.html.

Related Material