Trajectory-based vehicle tracking at low frame rates
作者：
Highlights：
• A new vehicle tracking method is proposed for an embedded traffic surveillance system.
• The proposed method demonstrates efficient tracking performance at a low frame rate.
• The proposed method employs greedy data association based on appearance and position similarities.
• To manage abrupt appearance changes, manifold learning is used.
• To manage abrupt motion changes, trajectory information is used to predict the next probable position.
摘要
•A new vehicle tracking method is proposed for an embedded traffic surveillance system.•The proposed method demonstrates efficient tracking performance at a low frame rate.•The proposed method employs greedy data association based on appearance and position similarities.•To manage abrupt appearance changes, manifold learning is used.•To manage abrupt motion changes, trajectory information is used to predict the next probable position.
论文关键词：Traffic surveillance,Trajectory identification,Vehicle detection and tracking,Data association,Appearance and position similarities
论文评审过程：Received 1 November 2016, Revised 10 March 2017, Accepted 11 March 2017, Available online 14 March 2017, Version of Record 11 April 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.03.023
