Multi–feature fusion tracking algorithm based on peak–context learning
作者：
Highlights：
• A robust tracking algorithm is proposed that imposes an elastic net regression.
• The optimization problem is improved by adding a contextual information constraint.
• A novel self-adaptive hand-crafted feature fusion strategy is proposed.
• An updating strategy is adopted in the tracking model using a response threshold.
• An efficient multi-scale mechanism is imposed in the tracker using a scale pool.
摘要
•A robust tracking algorithm is proposed that imposes an elastic net regression.•The optimization problem is improved by adding a contextual information constraint.•A novel self-adaptive hand-crafted feature fusion strategy is proposed.•An updating strategy is adopted in the tracking model using a response threshold.•An efficient multi-scale mechanism is imposed in the tracker using a scale pool.
论文关键词：Visual tracking,Correlation filter,Kernel trick,Elastic net,Context–aware,Feature fusion
论文评审过程：Received 6 February 2022, Revised 19 April 2022, Accepted 21 April 2022, Available online 26 April 2022, Version of Record 4 May 2022.
论文官网地址：https://doi.org/10.1016/j.imavis.2022.104468
