MCTrack: A Unified 3D Multi-Object Tracking Framework for Autonomous Driving

• DOI: 10.48550/arxiv.2409.16149
• Publication Date: 2024-09-23
• Authors: Xiyang Wang; Shouzheng Qi; Jia Zhao; Hangning Zhou; Siyu Zhang; Guoan Wang; Kai Tu; Songlin Guo; Jianbo Zhao; Jian Li; Mu Yang
• Keywords: Tracking (education)

This paper introduces MCTrack, a new 3D multi-object tracking method that achieves state-of-the-art (SOTA) performance across KITTI, nuScenes, and Waymo datasets. Addressing the gap in existing paradigms, which often perform well on specific datasets but lack generalizability, MCTrack offers unified solution. Additionally, we have standardized format of perceptual results various datasets, termed BaseVersion, facilitating researchers field (MOT) to concentrate core algorithmic development without undue burden data preprocessing. Finally, recognizing limitations current evaluation metrics, propose novel set assesses motion information output, such as velocity acceleration, crucial for downstream tasks. The source codes proposed are available at this link: https://github.com/megvii-research/MCTrack}{https://github.com/megvii-research/MCTrack