Abstract
Providing reliable uncertainty quantification for complex visual tasks such as object detection is of utmost importance for safety-critical applications such as autonomous driving, tumor detection, etc. Conformal prediction methods offer simple yet practical means to build uncertainty estimations that come with probabilistic guarantees. In this paper we apply such methods to the task of object localization and illustrate our analysis on a pedestrian detection use-case. We highlight both theoretical and practical implications of our analysis.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
All the coordinates of the true box will be found inside the rectangle defined by the predicted bounding box of the object.
- 2.
More complex variants exist. The typical process outlined here is more precisely known as split conformal prediction.
- 3.
The errors are sometimes called “residuals” (hence the \(R^i\) notation).
- 4.
Mathematically speaking, it is in fact sufficient that the calibration data and the data at inference time are exchangeable, conditionally on the training data.
- 5.
The \(1-\alpha \) guarantee only holds on average over all calibration sets, see Sect. 6.
- 6.
These coverage values include statistical error margins at level \(95\%\).
- 7.
References
Angelopoulos, A.N., Bates, S.: A gentle introduction to conformal prediction and distribution-free uncertainty quantification (2021). arXiv:2107.07511
Azevedo, T., de Jong, R., Maji, P.: Stochastic-YOLO: efficient probabilistic object detection under dataset shifts. In: ML4AD Workshop, NeurIPS 2020 (2020)
Barber, R.F., Candes, E.J., Ramdas, A., Tibshirani, R.J.: Conformal prediction beyond exchangeability (2022). arXiv:2202.13415
Bates, S., Angelopoulos, A., Lei, L., Malik, J., Jordan, M.I.: Distribution-free, risk-controlling prediction sets. J. ACM 68(6), 1–34 (2021)
Bickel, P.J., Doksum, K.A.: Mathematical Statistics: Basic Ideas and Selected Topics, vol. 1. Chapman and Hall/CRC, London (2015)
Bonnin, H., et al.: Can we reconcile safety objectives with machine learning performances? In: ERTS 2022 (2022)
Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S.: End-to-end object detection with transformers. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12346, pp. 213–229. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58452-8_13
Chih-Hong Cheng, T.S., Burton, S.: Logically sound arguments for the effectiveness of ML safety measures (2021). arXiv:2111.02649
Cordts, M., et al.: The cityscapes dataset for semantic urban scene understanding. In: CVPR 2016 (2016)
Deepshikha, K., Yelleni, S.H., Srijith, P.K., Mohan, C.K.: Monte Carlo DropBlock for modelling uncertainty in object detection (2021). arXiv:2108.03614
Ducoffe, M., Gerchinovitz, S., Sen Gupta, J.: A high-probability safety guarantee for shifted neural network surrogates. In: SafeAI 2020 (2020)
Feng, D., Harakeh, A., Waslander, S.L., Dietmayer, K.: A review and comparative study on probabilistic object detection in autonomous driving. IEEE T-ITS, 1–20 (2021)
Girshick, R.B.: Fast R-CNN. In: ICCV 2015 (2015)
Girshick, R.B., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: CVPR 2014 (2014)
Harakeh, A., Smart, M., Waslander, S.L.: BayesOD: a Bayesian approach for uncertainty estimation in deep object detectors. In: ICRA 2020 (2020)
Kendall, A., Gal, Y.: What uncertainties do we need in Bayesian deep learning for computer vision? In: NeurIPS 2017 (2017)
Kläs, M., Jöckel, L.: A framework for building uncertainty wrappers for AI/ML-based data-driven components. In: Casimiro, A., Ortmeier, F., Schoitsch, E., Bitsch, F., Ferreira, P. (eds.) SAFECOMP 2020. LNCS, vol. 12235, pp. 315–327. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-55583-2_23
Kläs, M., Sembach, L.: Uncertainty wrappers for data-driven models: increase the transparency of AI/ML-based models through enrichment with dependable situation-aware uncertainty estimates. In: WAISE 2019 (2019)
Kraus, F., Dietmayer, K.: Uncertainty estimation in one-stage object detection. In: ITSC 2019 (2019)
Kuhn, H.W.: The Hungarian method for the assignment problem. Naval Res. Logistics Q. 2(1–2), 83–97 (1955)
Lakshminarayanan, B., Pritzel, A., Blundell, C.: Simple and scalable predictive uncertainty estimation using deep ensembles. In: NeurIPS 2017 (2017)
Le, M.T., Diehl, F., Brunner, T., Knol, A.: Uncertainty estimation for deep neural object detectors in safety-critical applications. In: ITSC 2018 (2018)
Lei, J., G’Sell, M., Rinaldo, A., Tibshirani, R.J., Wasserman, L.: Distribution-free predictive inference for regression. JASA 113(523), 1094–1111 (2018)
Lin, T., Dollar, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: CVPR 2017 (2017)
Lin, T., Goyal, P., Girshick, R., He, K., Dollar, P.: Focal loss for dense object detection. IEEE T-PAMI 42(02), 318–327 (2020)
Lin, T.-Y., et al.: Microsoft COCO: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8693, pp. 740–755. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10602-1_48
Liu, W., et al.: SSD: single shot MultiBox detector. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9905, pp. 21–37. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46448-0_2
Lyu, Z., Gutierrez, N., Rajguru, A., Beksi, W.J.: Probabilistic object detection via deep ensembles. In: Bartoli, A., Fusiello, A. (eds.) ECCV 2020. LNCS, vol. 12540, pp. 67–75. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-65414-6_7
Miller, D., Dayoub, F., Milford, M., Sunderhauf, N.: Evaluating merging strategies for sampling-based uncertainty techniques in object detection. In: ICRA 2019 (2019)
Miller, D., Nicholson, L., Dayoub, F., Sünderhauf, N.: Dropout sampling for robust object detection in open-set conditions. In: ICRA 2018 (2018)
Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: CVPR 2016 (2016)
Redmon, J., Farhadi, A.: YOLO9000: better, faster, stronger. In: CVPR 2017 (2017)
Redmon, J., Farhadi, A.: YOLOv3: an incremental improvement (2018). arXiv:1804.02767
Rucklidge, W.: Efficiently locating objects using the hausdorff distance. IJCV 24, 251–270 (1997)
Tibshirani, R.J., Barber, R.F., Candes, E.J., Ramdas, A.: Conformal prediction under covariate shift. In: NeurIPS 2019 (2019)
Schuster, T., Seferis, E., Burton, S., Cheng, C.H.: Unaligned but safe - formally compensating performance limitations for imprecise 2D object detection (2022). arXiv:2202.05123
Vovk, V., Gammerman, A., Shafer, G.: Algorithmic Learning in a Random World. Springer, New York (2005). https://doi.org/10.1007/b106715
Yu, F., et al.: BDD100K: a diverse driving dataset for heterogeneous multitask learning (2018). arXiv:1805.04687
Acknowledgements
This work has benefited from the AI Interdisciplinary Institute ANITI, which is funded by the French “Investing for the Future - PIA3” program under the Grant agreement ANR-19-P3IA-0004. The authors gratefully acknowledge the support of the DEEL project (https://www.deel.ai/).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
de Grancey, F., Adam, JL., Alecu, L., Gerchinovitz, S., Mamalet, F., Vigouroux, D. (2022). Object Detection with Probabilistic Guarantees: A Conformal Prediction Approach. In: Trapp, M., Schoitsch, E., Guiochet, J., Bitsch, F. (eds) Computer Safety, Reliability, and Security. SAFECOMP 2022 Workshops . SAFECOMP 2022. Lecture Notes in Computer Science, vol 13415. Springer, Cham. https://doi.org/10.1007/978-3-031-14862-0_23
Download citation
DOI: https://doi.org/10.1007/978-3-031-14862-0_23
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-14861-3
Online ISBN: 978-3-031-14862-0
eBook Packages: Computer ScienceComputer Science (R0)