Consistent multi-layer subtask tracker via hyper-graph regularization

作者：

Highlights：

• Develop the online multi-subtask learning framework for robust object tracking with novel task definition.

• The relationships among and inside candidates or training samples are mined by hyper-graph regularization.

• Simultaneously learn and update the adaptively discriminative subspace and classifier.

• Consistent multi-subtask tracker is a general model for most existing multi-task trackers.

摘要

•Develop the online multi-subtask learning framework for robust object tracking with novel task definition.•The relationships among and inside candidates or training samples are mined by hyper-graph regularization.•Simultaneously learn and update the adaptively discriminative subspace and classifier.•Consistent multi-subtask tracker is a general model for most existing multi-task trackers.

论文关键词：Multi-layer subtask learning,Intrinsic geometrical structure,Graph regularization,Normalized collaborate metric,Object tracking

论文评审过程：Received 21 June 2016, Revised 29 January 2017, Accepted 3 February 2017, Available online 9 February 2017, Version of Record 1 March 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.02.008