Calibrating the Full Predictive Class Distribution of 3D Object Detectors for Autonomous Driving
This work addresses uncertainty estimation for safer autonomous driving, but it is incremental as it builds on existing calibration methods.
The paper tackles confidence calibration for 3D object detectors in autonomous driving by proposing auxiliary loss terms to calibrate the full predictive class distribution, finding that combining their full-class loss with isotonic regression improves calibration for CenterPoint and PillarNet, but not for DSVT-Pillar.
In autonomous systems, precise object detection and uncertainty estimation are critical for self-aware and safe operation. This work addresses confidence calibration for the classification task of 3D object detectors. We argue that it is necessary to regard the calibration of the full predictive confidence distribution over all classes and deduce a metric which captures the calibration of dominant and secondary class predictions. We propose two auxiliary regularizing loss terms which introduce either calibration of the dominant prediction or the full prediction vector as a training goal. We evaluate a range of post-hoc and train-time methods for CenterPoint, PillarNet and DSVT-Pillar and find that combining our loss term, which regularizes for calibration of the full class prediction, and isotonic regression lead to the best calibration of CenterPoint and PillarNet with respect to both dominant and secondary class predictions. We further find that DSVT-Pillar can not be jointly calibrated for dominant and secondary predictions using the same method.