Which Framework is Suitable for Online 3D Multi-Object Tracking for Autonomous Driving with Automotive 4D Imaging Radar?

TL;DR

This paper systematically compares different frameworks for online 3D multi-object tracking using automotive 4D imaging radar, highlighting the strengths and weaknesses of each approach in real-world autonomous driving scenarios.

Contribution

It provides the first comprehensive benchmark and analysis of point object tracking, joint-detection-and-tracking, and proposed hybrid frameworks for 4D radar-based 3D MOT.

Findings

TBD-POT remains most effective for high performance and low complexity

TBD-EOT can outperform TBD-POT in certain situations

JDT-EOT faces significant challenges and performs inadequately

Abstract

Online 3D multi-object tracking (MOT) has recently received significant research interests due to the expanding demand of 3D perception in advanced driver assistance systems (ADAS) and autonomous driving (AD). Among the existing 3D MOT frameworks for ADAS and AD, conventional point object tracking (POT) framework using the tracking-by-detection (TBD) strategy has been well studied and accepted for LiDAR and 4D imaging radar point clouds. In contrast, extended object tracking (EOT), another important framework which accepts the joint-detection-and-tracking (JDT) strategy, has rarely been explored for online 3D MOT applications. This paper provides the first systematic investigation of the EOT framework for online 3D MOT in real-world ADAS and AD scenarios. Specifically, the widely accepted TBD-POT framework, the recently investigated JDT-EOT framework, and our proposed TBD-EOT framework are compared via extensive evaluations on two open source 4D imaging radar datasets: View-of-Delft and TJ4DRadSet. Experiment results demonstrate that the conventional TBD-POT framework remains preferable for online 3D MOT with high tracking performance and low computational complexity, while the proposed TBD-EOT framework has the potential to outperform it in certain situations. However, the results also show that the JDT-EOT framework encounters multiple problems and performs inadequately in evaluation scenarios. After analyzing the causes of these phenomena based on various evaluation metrics and visualizations, we provide possible guidelines to improve the performance of these MOT frameworks on real-world data. These provide the first benchmark and important insights for the future development of 4D imaging radar-based online 3D MOT algorithms.