Inferring Spatial Uncertainty in Object Detection

Abstract

The availability of real-world datasets is the prerequisite for developing object detection methods for autonomous driving. While ambiguity exists in object labels due to error-prone annotation process or sensor observation noises, current object detection datasets only provide deterministic annotations without considering their uncertainty. This precludes an in-depth evaluation among different object detection methods, especially for those that explicitly model predictive probability. In this work, we propose a generative model to estimate bounding box label uncertainties from LiDAR point clouds, and define a new representation of the probabilistic bounding box through spatial distribution. Comprehensive experiments show that the proposed model represents uncertainties commonly seen in driving scenarios. Based on the spatial distribution, we further propose an extension of IoU, called the Jaccard IoU (JIoU), as a new evaluation metric that incorporates label uncertainty. Experiments on the KITTI and the Waymo Open Datasets show that JIoU is superior to IoU when evaluating probabilistic object detectors.

References in corpus (5)

• Training Convolutional Networks with Noisy Labels
• Scalability in Perception for Autonomous Driving: Waymo Open Dataset
• Augmented LiDAR Simulator for Autonomous Driving
• Can We Trust You? On Calibration of a Probabilistic Object Detector for Autonomous Driving
• Towards Better Performance and More Explainable Uncertainty for 3D Object Detection of Autonomous Vehicles

Cited by in corpus (4)

• Semantics for Robotic Mapping, Perception and Interaction: A Survey
• A Review of Testing Object-Based Environment Perception for Safe Automated Driving
• Uncertainty for Identifying Open-Set Errors in Visual Object Detection
• Labels Are Not Perfect: Improving Probabilistic Object Detection via Label Uncertainty