Aberrance suppressed spatio-temporal correlation filters for visual object tracking
作者：
Highlights：
• Proposed a tracking algorithm concerning the spatial information and aberrance repressed-based correlation filter.
• The proposed method utilizes the spatial and temporal information of the target to avoid the boundary effects and drastic appearance changes.
• Introduces the aberrance term into proposed tracker to avoid tracking failure caused by sudden changes appeared in between the two frames.
• Provides an optimal solution for the model updating based on the ADMM method.
• The experimental evaluations and comparisons with existing trackers are provided by following benchmark datasets: OTB2013, OTB2015, TempleColor128 and UAV123.
摘要
•Proposed a tracking algorithm concerning the spatial information and aberrance repressed-based correlation filter.•The proposed method utilizes the spatial and temporal information of the target to avoid the boundary effects and drastic appearance changes.•Introduces the aberrance term into proposed tracker to avoid tracking failure caused by sudden changes appeared in between the two frames.•Provides an optimal solution for the model updating based on the ADMM method.•The experimental evaluations and comparisons with existing trackers are provided by following benchmark datasets: OTB2013, OTB2015, TempleColor128 and UAV123.
论文关键词：Visual object tracking,Correlation filter,Spatio-temporal information,Radical changes
论文评审过程：Received 9 March 2020, Revised 17 February 2021, Accepted 23 February 2021, Available online 27 February 2021, Version of Record 3 March 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.107922
