Visual object tracking with multi-scale superpixels and color-feature guided kernelized correlation filters
作者：
Highlights：
•
• Tracking is treated as optimizing problem, KCF is embedded into Bayesian framework.
• Multi-scale superpixel and color feature guided method for filtering confidence map.
• Min–max criterion and center distance matrix consider structure information as well.
• Monitoring update strategy for monitoring and adjusting tracking dynamically.
• PSR, tPSR plots and a self-made dataset are proposed to evaluate the performance.
摘要
•Tracking is treated as optimizing problem, KCF is embedded into Bayesian framework.•Multi-scale superpixel and color feature guided method for filtering confidence map.•Min–max criterion and center distance matrix consider structure information as well.•Monitoring update strategy for monitoring and adjusting tracking dynamically.•PSR, tPSR plots and a self-made dataset are proposed to evaluate the performance.
论文关键词：Visual tracking,Superpixel,Kernelized correlation filter,Bayesian filter,Color feature guided confidence map,Multi-scale superpixels
论文评审过程：Received 13 July 2017, Revised 23 January 2018, Accepted 23 January 2018, Available online 1 February 2018, Version of Record 7 February 2018.
论文官网地址：https://doi.org/10.1016/j.image.2018.01.005
