提升自动驾驶3D目标检测的置信度校准，让模型更懂自己的不确定性
Calibrating the Full Predictive Class Distribution of 3D Object Detectors for Autonomous Driving
- 提出两种正则化损失，训练时直接优化全类别置信度分布
- 结合等熵回归后，中心点和柱状网络在主类与次类预测上均实现最优校准
- 发现DSVT-Pillar难以同时校准主次类别，提示模型结构影响校准效果
在自动驾驶系统中，精确的目标检测与不确定性估计对自知性与安全性至关重要。本文聚焦于3D目标检测分类任务的置信度校准问题。我们认为，需对所有类别上的完整预测置信度分布进行校准，并构建能同时捕捉主类与次类预测校准程度的指标。为此，提出两种辅助正则化损失项，分别以主类预测或全预测向量的校准为目标进行训练。我们在CenterPoint、PillarNet和DSVT-Pillar上评估多种事后与训练时校准方法，发现结合全预测校准损失项与等熵回归，可使CenterPoint和PillarNet在主类与次类预测上均达到最佳校准效果。此外，我们发现DSVT-Pillar无法使用同一方法同时校准主类与次类预测。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。