Object tracking using discriminative sparse appearance model
作者：
Highlights：
• A dictionary is learned by comprehensively considering both representation capability and discriminative strength.
• The target appearance is modeled as a discriminative sparse model.
• A similarity coefficient is defined to measure the similarity between the target model and the candidate.
• An update strategy is proposed to reduce the adverse effects caused by appearance changes.
摘要
Highlights•A dictionary is learned by comprehensively considering both representation capability and discriminative strength.•The target appearance is modeled as a discriminative sparse model.•A similarity coefficient is defined to measure the similarity between the target model and the candidate.•An update strategy is proposed to reduce the adverse effects caused by appearance changes.
论文关键词：Visual tracking,Sparse representation,Dictionary learning,Adaptive update,Bayesian inference framework
论文评审过程：Received 27 January 2015, Revised 13 May 2015, Accepted 29 June 2015, Available online 16 July 2015, Version of Record 31 July 2015.
论文官网地址：https://doi.org/10.1016/j.image.2015.06.012
