Robust vision tracking by online random ferns and template library
作者：
Highlights：
• We propose a robust tracking algorithm based on random ferns and template library.
• Random Gaussian difference is adopted to generate binary features.
• Semi-naive Bayes based random ferns are used to establish discriminative model.
• Co-training of discriminative model and template library improves the accuracy.
• Experiments show a good performance of our tracker on challenging sequences.
摘要
Highlights•We propose a robust tracking algorithm based on random ferns and template library.•Random Gaussian difference is adopted to generate binary features.•Semi-naive Bayes based random ferns are used to establish discriminative model.•Co-training of discriminative model and template library improves the accuracy.•Experiments show a good performance of our tracker on challenging sequences.
论文关键词：Vision tracking,Random ferns,Template library,Discriminative and generative models
论文评审过程：Received 6 August 2013, Revised 5 January 2014, Accepted 14 March 2014, Available online 4 April 2014.
论文官网地址：https://doi.org/10.1016/j.image.2014.03.001
