End-to-End Joint Multi-Object Detection and Tracking for Intelligent Transportation Systems

• Published in: Chinese journal of mechanical engineering Vol. 36; no. 1; pp. 138 - 11
• Main Authors: Xu, Qing, Lin, Xuewu, Cai, Mengchi, Guo, Yu-ang, Zhang, Chuang, Li, Kai, Li, Keqiang, Wang, Jianqiang, Cao, Dongpu
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 20-11-2023; Springer Nature B.V; SpringerOpen
• Edition: English ed.
• Subjects: Advanced Transportation Equipment; Algorithms; Electrical Machines and Networks; Electronics and Microelectronics; End-to-end tracking; Engineering; Engineering Thermodynamics; Feature extraction; Global correlation network; Heat and Mass Transfer; Instrumentation; Intelligent transportation systems; Joint detection and tracking; Machines; Manufacturing; Mechanical Engineering; Multiple target tracking; Object recognition; Original Article; Power Electronics; Processes; Theoretical and Applied Mechanics; Transportation networks; End-to-end tracking; Global correlation network; Intelligent transportation systems; Joint detection and tracking
• ISSN: 2192-8258; 1000-9345; 2192-8258
• DOI: 10.1186/s10033-023-00962-x

Environment perception is one of the most critical technology of intelligent transportation systems (ITS). Motion interaction between multiple vehicles in ITS makes it important to perform multi-object tracking (MOT). However, most existing MOT algorithms follow the tracking-by-detection framework, which separates detection and tracking into two independent segments and limit the global efficiency. Recently, a few algorithms have combined feature extraction into one network; however, the tracking portion continues to rely on data association, and requires complex post-processing for life cycle management. Those methods do not combine detection and tracking efficiently. This paper presents a novel network to realize joint multi-object detection and tracking in an end-to-end manner for ITS, named as global correlation network (GCNet). Unlike most object detection methods, GCNet introduces a global correlation layer for regression of absolute size and coordinates of bounding boxes, instead of offsetting predictions. The pipeline of detection and tracking in GCNet is conceptually simple, and does not require complicated tracking strategies such as non-maximum suppression and data association. GCNet was evaluated on a multi-vehicle tracking dataset, UA-DETRAC, demonstrating promising performance compared to state-of-the-art detectors and trackers.