FAST3D: Flow-Aware Self-Training for 3D Object Detectors

TL;DR

This paper introduces FAST3D, a flow-aware self-training method that leverages scene flow for unsupervised domain adaptation in 3D object detection, significantly improving performance on continuous LiDAR data without prior target domain knowledge.

Contribution

It proposes a novel flow-based self-training approach that uses scene flow to generate reliable pseudo-labels for unsupervised domain adaptation in 3D object detection.

Findings

Significant performance improvement over state-of-the-art methods.

Effective pseudo-label generation using scene flow and flow consistency.

Successful adaptation on the Waymo dataset without target domain labels.

Abstract

In the field of autonomous driving, self-training is widely applied to mitigate distribution shifts in LiDAR-based 3D object detectors. This eliminates the need for expensive, high-quality labels whenever the environment changes (e.g., geographic location, sensor setup, weather condition). State-of-the-art self-training approaches, however, mostly ignore the temporal nature of autonomous driving data. To address this issue, we propose a flow-aware self-training method that enables unsupervised domain adaptation for 3D object detectors on continuous LiDAR point clouds. In order to get reliable pseudo-labels, we leverage scene flow to propagate detections through time. In particular, we introduce a flow-based multi-target tracker, that exploits flow consistency to filter and refine resulting tracks. The emerged precise pseudo-labels then serve as a basis for model re-training. Starting with a pre-trained KITTI model, we conduct experiments on the challenging Waymo Open Dataset to demonstrate the effectiveness of our approach. Without any prior target domain knowledge, our results show a significant improvement over the state-of-the-art.