Visual tracking via adaptive multi-task feature learning with calibration and identification

作者：

Highlights：

• Multi-task feature learning model with calibration and identification is proposed.

• The model calibrates loss function and identifies outlier tasks to select features.

• The model can effectively applied in object tracking.

摘要

•Multi-task feature learning model with calibration and identification is proposed.•The model calibrates loss function and identifies outlier tasks to select features.•The model can effectively applied in object tracking.

论文关键词：Multi-task feature learning,Object tracking,Feature selection,Sparse representation

论文评审过程：Received 26 October 2015, Revised 28 September 2016, Accepted 29 September 2016, Available online 3 October 2016, Version of Record 11 October 2016.

论文官网地址：https://doi.org/10.1016/j.image.2016.09.009