Computer Engineering and Applications ›› 2020, Vol. 56 ›› Issue (9): 13-23.DOI: 10.3778/j.issn.1002-8331.1911-0176

Previous Articles     Next Articles

Review of Visual Object Tracking Algorithms of Adaptive Direction and Scale

SHAN Yugang, HU Weiguo   

  1. 1.Education Institute, Hubei University of Arts and Science, Xiangyang, Hubei 441053, China
    2.A Certain Unit of the PLA, China
  • Online:2020-05-01 Published:2020-04-29



  1. 1.湖北文理学院 教育学院,湖北 襄阳 441053
    2.中国人民解放军 某部队


The problem of object scale and direction change has always been a difficulty in the process of visual object tracking. How to deal with the change of object scale and direction effectively is an important factor to ensure the robustness of object tracking algorithm. This paper introduces the development of video object tracking, and classifies the existing object scale and direction tracking algorithms:incremental search, mean shift iteration, corner matching, region second-order moment, particle filter, correlation filter and depth learning tracking algorithm. The basic idea of various algorithms and their processing methods of scale and direction are described, focusing on the analysis of the method of using depth learning technology to deal with the change of object scale and direction, analyzing the advantages and disadvantages of various algorithms, and pointing out their applicable occasions. The future development trend of object scale and direction tracking is prospected, and the main challenges and problems are put forward, which can be used for reference for related researchers.

Key words: object tracking, scale and direction, adaptation, visual



关键词: 目标跟踪, 尺度方向, 自适应, 视觉