Instance-Level Post Hoc Uncertainty Quantification in Object Detection
2026-06-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionArtificial Intelligence
AI summaryⓘ
The authors focus on improving how self-driving cars can estimate their confidence in detecting objects like cars or pedestrians. They want to measure uncertainty without retraining the entire detection system, which is important for real-world use. To do this, they create a method called Monte-Carlo generalized linearized model (MC-GLM) that gives quick and instance-specific uncertainty estimates. Their tests show that this approach works well and is efficient when used on a common driving dataset.
object detectionautonomous drivinguncertainty quantificationLaplace approximationMonte Carlo methodsbounding boxpost hoc analysisCenterPoint detectornuScenes dataset
Authors
Chongzhe Zhang, Zifan Zeng, Qunli Zhang, Feng Liu, Zheng Hu
Abstract
Object detection is a safety-critical component of autonomous driving. It is essential to quantify the uncertainty in bounding-box predictions for safety assurance. Post hoc uncertainty quantification without retraining aligns with real-world deployment requirements; therefore, we employ the Laplace approximation. Because instance-level uncertainty is needed, linearized inference methods that require multiple backpropagations are not time-efficient, and sampling-based methods are not fully post hoc. We propose Monte-Carlo generalized linearized model (MC-GLM), which provides instance-level and approximately post hoc uncertainty quantification. The number of samples required in the Monte Carlo step is constant and independent of the number of output instances, so it can be parallelized. Experiments on the nuScenes dataset with the CenterPoint detector validate the effectiveness of our method, and the resulting uncertainties exhibit good quality.