Uncertainty-aware Mean Teacher for Source-free Unsupervised Domain Adaptive 3D Object Detection

Abstract

Pseudo-label based self training approaches are a popular method for source-free unsupervised domain adaptation. However, their efficacy depends on the quality of the labels generated by the source trained model. These labels may be incorrect with high confidence, rendering thresholding methods ineffective. In order to avoid reinforcing errors caused by label noise, we propose an uncertainty-aware mean teacher framework which implicitly filters incorrect pseudo-labels during training. Leveraging model uncertainty allows the mean teacher network to perform implicit filtering by down-weighing losses corresponding uncertain pseudo-labels. Effectively, we perform automatic soft-sampling of pseudo-labeled data while aligning predictions from the student and teacher networks. We demonstrate our method on several domain adaptation scenarios, from cross-dataset to cross-weather conditions, and achieve state-of-the-art performance in these cases, on the KITTI lidar target dataset.

References in corpus (5)

• IPOD: Intensive Point-based Object Detector for Point Cloud
• PointDAN: A Multi-Scale 3D Domain Adaption Network for Point Cloud Representation
• Lidar Light Scattering Augmentation (LISA): Physics-based Simulation of Adverse Weather Conditions for 3D Object Detection
• Unbiased Teacher for Semi-Supervised Object Detection
• Pseudo-labeling for Scalable 3D Object Detection