DA-SACOT: Domain adaptive-segmentation guided attention for correlation based object tracking
作者：
Highlights：
• Segmentation guided attention for single object tracking.
• Eliminates model drift and enables long term tracking.
• We develop two variants of the proposed segmentation guided tracking.
• A domain adaptation technique is integrated to incorporate target specific knowledge.
• An initial offline fine-tuning and online model update technique is developed.
摘要
•Segmentation guided attention for single object tracking.•Eliminates model drift and enables long term tracking.•We develop two variants of the proposed segmentation guided tracking.•A domain adaptation technique is integrated to incorporate target specific knowledge.•An initial offline fine-tuning and online model update technique is developed.
论文关键词：Visual object tracking,Object segmentation,Online learning,Domain adaptation,Correlation filter based tracker
论文评审过程：Received 2 September 2020, Revised 11 May 2021, Accepted 15 May 2021, Available online 21 May 2021, Version of Record 5 June 2021.
论文官网地址：https://doi.org/10.1016/j.imavis.2021.104215
