mlsanityData to 7 Oct 2026

Can We Trust You? On Calibration of a Probabilistic Object Detector for Autonomous Driving

arXiv 1909.12358 ↗First posted 26 Sep 2019cs.RO cs.CVcited by 30

Not on the map. Why?

Abstract (arXiv)

Reliable uncertainty estimation is crucial for perception systems in safe autonomous driving. Recently, many methods have been proposed to model uncertainties in deep learning based object detectors. However, the estimated probabilities are often uncalibrated, which may lead to severe problems in safety critical scenarios. In this work, we identify such uncertainty miscalibration problems in a probabilistic LiDAR 3D object detection network, and propose three practical methods to significantly reduce errors in uncertainty calibration. Extensive experiments on several datasets show that our methods produce well-calibrated uncertainties, and generalize well between different datasets.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). First posted on arXiv 26 Sep 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 3; 2023: 5; 2024: 5; 2025: 3; 2026: 7.

30 papers.

Sort:

By area:

  1. 2021Estimating and Evaluating Regression Predictive Uncertainty in Deep Object Detectorscited by 37
  2. 2021How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Reviewcited by 36
  3. 2021Uncertainty for Identifying Open-Set Errors in Visual Object Detectioncited by 26
  4. 2023A Benchmark Study on Calibrationcited by 10
  5. 2021A Review of Testing Object-Based Environment Perception for Safe Automated Drivingcited by 9
  6. 2021From Black-box to White-box: Examining Confidence Calibration under different Conditionscited by 8
  7. 2022Parametric and Multivariate Uncertainty Calibration for Regression and Object Detectioncited by 8
  8. 2024Feature Clipping for Uncertainty Calibrationcited by 8
  9. 2021Bayesian Confidence Calibration for Epistemic Uncertainty Modellingcited by 5
  10. 2022LiDAR-MIMO: Efficient Uncertainty Estimation for LiDAR-based 3D Object Detectioncited by 5
  11. 2023Overcoming the Limitations of Localization Uncertainty: Efficient & Exact Non-Linear Post-Processing and Calibrationcited by 5
  12. 2021f-Cal: Calibrated aleatoric uncertainty estimation from neural networks for robot perceptioncited by 3
18 more
  1. 2022Calibrating Perception Uncertainty for Autonomous Drivingcited by 3
  2. 2024Towards Calibrated Deep Clustering Networkcited by 3
  3. 2024Run-time Monitoring of 3D Object Detection in Automated Driving Systems Using Early Layer Neural Activation Patternscited by 3
  4. 2024Consistency Calibration: Improving Uncertainty Calibration via Consistency among Perturbed Neighborscited by 3
  5. 2025Sample Margin-Aware Recalibration of Temperature Scalingcited by 3
  6. 2026Scalable Variational Bayesian Fine-Tuning of LLMs via Orthogonalized Low-Rank Adapterscited by 2
  7. 2026Low-Rank Adaptation Redux for Large Modelscited by 2
  8. 2023Perception and Semantic Aware Regularization for Sequential Confidence Calibrationcited by 1
  9. 2023A Synthetic Benchmarking Pipeline to Compare Camera Calibration Algorithmscited by 1
  10. 2025How Safe Will I Be Given What I Saw? Calibrated Prediction of Safety Chances for Image-Controlled Autonomycited by 1
  11. 2026Self-Aware Object Detection via Degradation Manifoldscited by 1
  12. 2026Query2Uncertainty: Robust Uncertainty Quantification and Calibration for 3D Object Detection under Distribution Shiftcited by 1
  13. 2023Probabilistic 3d regression with projected huber distributioncited by 0
  14. 2024Integrity Monitoring of 3D Object Detection in Automated Driving Systems using Raw Activation Patterns and Spatial Filteringcited by 0
  15. 2025Calibrating the Full Predictive Class Distribution of 3D Object Detectors for Autonomous Drivingcited by 0
  16. 2026Towards Trustworthy and Explainable AI for Perception Models: From Concept to Prototype Vehicle Deploymentcited by 0
  17. 2026Calibrating Probabilistic Object Detectors with Annotator Disagreementcited by 0
  18. 2026Instance-Level Post Hoc Uncertainty Quantification in Object Detectioncited by 0