Visual tracking with multiple Hough detectors
作者：
Highlights：
• The Hough detector based on random ferns with the 5 bit Binary Patterns is proposed.
• The detectors are created automatically by the feature selection.
• All the detectors are used to locate the target collaboratively in a continuous and selective detection way.
• The tracking framework is simple and easy to implement.
• The tracker outperforms or achieves comparable results to other methods on the CVPR2013 Visual Tracker Benchmark.
摘要
•The Hough detector based on random ferns with the 5 bit Binary Patterns is proposed.•The detectors are created automatically by the feature selection.•All the detectors are used to locate the target collaboratively in a continuous and selective detection way.•The tracking framework is simple and easy to implement.•The tracker outperforms or achieves comparable results to other methods on the CVPR2013 Visual Tracker Benchmark.
论文关键词：Visual tracking,Multiple Hough detectors,Feature selection,Detector selection
论文评审过程：Received 27 June 2016, Revised 3 July 2017, Accepted 7 August 2017, Available online 15 August 2017, Version of Record 22 September 2017.
论文官网地址：https://doi.org/10.1016/j.imavis.2017.08.001
