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

Multi-object tracking (MOT) is among crucial applications in modern advanced driver assistance systems (ADAS) and autonomous driving (AD) systems. Global nearest neighbor (GNN) filter, as the earliest random vector Bayesian tracking framework, has been adopted in most of state-of-the-arts trackers and widely accepted in the automotive industry. With the development of random finite set (RFS) theory, the RFS Bayesian filters have been applied in MOT tasks recently. However, their usefulness in the real traffic for ADAS and AD application is still open to doubt. In this paper, we firstly demonstrate the latest RFS Bayesian tracking framework could be superior to typical random vector Bayesian tracking framework like GNN, via a systematic comparative study of both traditional random vector Bayesian filters with rule-based heuristic track maintenance and RFS Bayesian filters on nuScenes validation dataset. Then, we propose a RFS-based tracker, namely Poisson multi-Bernoulli filter using the global nearest neighbor (GNN-PMB), for LiDAR-based MOT tasks. This GNN-PMB tracker is simple to use but can achieve competitive results on nuScenes dataset. Specifically, the proposed GNN-PMB tracker outperforms most of the state-of-the-art LiDAR-only trackers and LiDAR and camera fusion-based trackers, ranks the 3rd among all LiDAR-only trackers on nuScenes tracking task leader board1 at the time of submission. Our code is available at https://github.com/chisyliu/gnn pmb tracker.