Online parameter tuning for object tracking algorithms
作者：
Highlights：
• We present a new control approach to adapt trackers to scene condition variations.
• Tracking context is defined as six features describing scene condition.
• Best tracker parameters are learned offline for tracking contexts.
• Trackers are then controlled by tuning online their parameters.
• Experimental results are compared with several recent state of the art trackers.
摘要
•We present a new control approach to adapt trackers to scene condition variations.•Tracking context is defined as six features describing scene condition.•Best tracker parameters are learned offline for tracking contexts.•Trackers are then controlled by tuning online their parameters.•Experimental results are compared with several recent state of the art trackers.
论文关键词：Object tracking,Online parameter tuning,Controller,Self-adaptation,Machine learning
论文评审过程：Received 28 October 2012, Revised 20 January 2014, Accepted 13 February 2014, Available online 21 February 2014.
论文官网地址：https://doi.org/10.1016/j.imavis.2014.02.003
