GNN-PMB: A Simple but Effective Online 3D Multi-Object Tracker without Bells and Whistles

Liu, Jianan, Bai, Liping, Xia, Yuxuan, Huang, Tao, Zhu, Bing, and Han, Qing-Long (2022) GNN-PMB: A Simple but Effective Online 3D Multi-Object Tracker without Bells and Whistles. IEEE Transactions on Intelligent Vehicles. (In Press)

[img] PDF (Author Accepted Version) - Accepted Version
Restricted to Repository staff only

 
3


Abstract

Multi-object tracking (MOT) is among crucial applications in modern advanced driver assistance systems (ADAS) and autonomous driving (AD) systems. The global nearest neighbor (GNN) filter, as the earliest random vector-based Bayesian tracking framework, has been adopted in most of state-of-the-arts trackers in the automotive industry. The development of random finite set (RFS) theory facilitates a mathematically rigorous treatment of the MOT problem, and different variants of RFS-based Bayesian filters have then been proposed. However, their effectiveness in the real ADAS and AD application is still an open problem. In this paper, it is demonstrated that the latest RFS-based Bayesian tracking framework could be superior to typical random vector-based Bayesian tracking framework via a systematic comparative study of both traditional random vector-based Bayesian filters with rule-based heuristic track maintenance and RFS-based Bayesian filters on the nuScenes validation dataset. An RFS-based tracker, namely Poisson multi-Bernoulli filter using the global nearest neighbor (GNN-PMB), is proposed to LiDAR-based MOT tasks. This GNN-PMB tracker is simple to use, and it achieves competitive results on the nuScenes dataset. Specifically, the proposed GNN-PMB tracker outperforms most state-of-the-art LiDAR-only trackers and LiDAR and camera fusion-based trackers, ranking the 3rd among all LiDAR-only trackers on nuScenes 3D tracking challenge leaderboard (https://bit.ly/3bQJ2CP) at the time of submission. Our code is available at https://github.com/chisyliu/GnnPmbTracker.

Item ID: 76650
Item Type: Article (Research - C1)
ISSN: 2379-8904
Keywords: Multi-object tracking, random vector-based Bayesian filters, random finite set-based Bayesian filters, GNNPMB, LiDAR, autonomous driving
Copyright Information: © 2022 IEEE.
Funders: Australian Research Council (ARC)
Projects and Grants: ARC DP220101634
Date Deposited: 30 Oct 2022 23:38
FoR Codes: 40 ENGINEERING > 4007 Control engineering, mechatronics and robotics > 400703 Autonomous vehicle systems @ 40%
46 INFORMATION AND COMPUTING SCIENCES > 4603 Computer vision and multimedia computation > 460301 Active sensing @ 30%
40 ENGINEERING > 4006 Communications engineering > 400607 Signal processing @ 30%
SEO Codes: 22 INFORMATION AND COMMUNICATION SERVICES > 2204 Information systems, technologies and services > 220408 Information systems @ 40%
24 MANUFACTURING > 2415 Transport equipment > 241502 Automotive equipment @ 60%
Downloads: Total: 3
Last 12 Months: 1
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page