J Shanghai Jiaotong Univ Sci ›› 2021, Vol. 26 ›› Issue (5): 598-606.doi: 10.1007/s12204-021-2349-6

• Intelligent Connected Vehicle • Previous Articles     Next Articles

Efficient Online Vehicle Tracking for Real–Virtual Mapping Systems

CHEN Jiacheng1 (陈佳诚), LI Lin2 (李 霖), YANG Xubo1 (杨旭波)   

  1. (1. School of Software, Shanghai Jiao Tong University, Shanghai 200240, China;2. Shanghai International Automobile City (Group) Co., Ltd., Shanghai 201805, China)
  • Received:2021-02-08 Online:2021-10-28 Published:2021-10-28

Abstract: Multi-object tracking is a vital problem as many applications require better tracking approaches. Although learning-based detectors are becoming extremely powerful, there are few tracking methods designed to work with them in real time. We explored an effcient fiexible online vehicle tracking-by-detection framework suitable for real–virtual mapping systems, which combines a non-recursive temporal window search with delayed output and produces stable trajectories despite noisy detection responses. Its computation speed meets the real-time requirements, whereas its performance is comparable with that of state-of-the-art online trackers on the DETRAC dataset. The trajectories from our approach also contain the target class and color information important for virtual vehicle motion reconstruction.

Key words: multi-object tracking, vehicle tracking, tracklet association, real–virtual mapping

CLC Number: