Real-time and robust object tracking in video via low-rank coherency analysis in feature space
作者：
Highlights：
• We propose a versatile, real-time, and robust video object tracking method.
• We define an efficient discriminative appearance model based on compressive sensing in a low dimensional feature space.
• We propose a novel low-rank decomposition based coherency analysis model for tracking and updating.
• We formulate a series of sparsity-measuring based criteria to handle various challenges of object tracking.
摘要
Highlights•We propose a versatile, real-time, and robust video object tracking method.•We define an efficient discriminative appearance model based on compressive sensing in a low dimensional feature space.•We propose a novel low-rank decomposition based coherency analysis model for tracking and updating.•We formulate a series of sparsity-measuring based criteria to handle various challenges of object tracking.
论文关键词：Localized compressive sensing representation,Fast lowrank approximation,Lowrank coherency tracking,Visual tracking
论文评审过程：Received 30 July 2014, Revised 7 November 2014, Accepted 27 January 2015, Available online 13 February 2015, Version of Record 16 May 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.01.025
