Laser & Optoelectronics Progress, Volume. 58, Issue 16, 1610011(2021)

Object Tracking Based on Adaptive Feature Fusion and Context-Aware

Yuanfa Ji1,2, Chuanji He1,2, Xiyan Sun1,2、*, and Ning Guo1,2
Author Affiliations
  • 1Guangxi Key Laboratory of Precision Navigation Technology and Application, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China
  • 2National & Local Joint Engineering Research Center of Satellite Navigation and Location Service, Guilin, Guangxi 541004, China;
  • show less
    Figures & Tables(10)
    Flowchart of the proposed algorithm
    APCE value and response value at different scenes. (a) Scene 1; (b) scene 2
    Color histogram and response map. (a) Original image; (b) color histogram; (c) response map of color feature
    Adaptive weight values for deep and shallow scores
    Comparison of tracking performance of three algorithms. (a)Precision; (b)success rate
    Comparison of the precision and success rate of 10 tracking algorithms on OTB-2013 benchmark. (a) Precision; (b) success rate
    Comparison of tracking precision of 8 attribute sequences on OTB-2013 benchmark
    Comparison of tracking success rates of 8 attribute sequences on OTB-2013 benchmark
    Qualitative comparison of 10 tracking algorithms on OTB-2013 benchmark
    • Table 1. Tracking speed of 10 tracking algorithms

      View table

      Table 1. Tracking speed of 10 tracking algorithms

      ParameterOURHCFSRDCFStapleLCTSiamFCRPTSAMFKCFDSST
      Tracking speed /(frame·s-1)51047528556817221
    Tools

    Get Citation

    Copy Citation Text

    Yuanfa Ji, Chuanji He, Xiyan Sun, Ning Guo. Object Tracking Based on Adaptive Feature Fusion and Context-Aware[J]. Laser & Optoelectronics Progress, 2021, 58(16): 1610011

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Image Processing

    Received: Jul. 31, 2020

    Accepted: Dec. 22, 2020

    Published Online: Aug. 16, 2021

    The Author Email: Xiyan Sun (sunxiyan1@163.com)

    DOI:10.3788/LOP202158.1610011

    Topics