Acta Optica Sinica, Volume. 37, Issue 9, 0915005(2017)

Multiple Feature Fusion based on Covariance Matrix for Visual Tracking

Zefenfen Jin*, Zhiqiang Hou, Wangsheng Yu, and Xin Wang
Author Affiliations
  • Information and Navigation College, Air Force Engineering University of PLA, Xi'an, Shaanxi 710077, China
  • show less
    Figures & Tables(12)
    Integral image
    Comparison of tracking results of different features. (a) The 649th frame of Basketball sequence; (b) the 873rd frame of Liquor sequence
    Population generation method of the ith frame
    Qualitative comparison of eight tracking algorithms. (a) Basketball; (b) Bolt; (c) David3; (d) Football1; (e) Jumping; (f) Liquor; (g) Matrix; (h) Skiing; (i) Ironman; (j) Jogging1; (k) Lemming; (l) MotorRolling
    Center position error curves. (a) Basketball; (b) Bolt; (c) David3; (d) Football1; (e) Jumping; (f) Liquor; (g) Matrix; (h) Skiing; (i) Ironman; (j) Jogging1; (k) Lemming; (l) MotorRolling
    Overlap rate curves. (a) Basketball; (b) Bolt; (c) David3; (d) Football1; (e) Jumping; (f) Liquor; (g) Matrix; (h) Skiing; (i) Ironman; (j) Jogging1; (k) Lemming; (l) MotorRolling
    Curves of precision and success rate. (a) Precision; (b) success rate
    Comparison of tracking precision under 11 scenes
    Comparison of tracking success rate under 11 scenes
    • Table 1. Generative rules of rotation angle of quantum rotation gate

      View table

      Table 1. Generative rules of rotation angle of quantum rotation gate

      xibif(x)≥f(b)Δθis(αiβi)
      αiβi>0αiβi<0αi=0βi=0
      00False00000
      00True00000
      01False00000
      01True0.05π-1+1±10
      10False0.01π-1+1±10
      10True0.025π+1-10±1
      11False0.005π+1-10±1
      11True0.025π+1-10±1
    • Table 2. Comparison of coverage rate of tracking results%

      View table

      Table 2. Comparison of coverage rate of tracking results%

      NameDLTTLDStruckASLADSSTSSTMILProposed algorithm
      Basketball53.92.39.9353.914.322.227.590.1
      Bolt4.291.41.711.4100.01.11.191.6
      David332.910.733.7351.654.035.768.378.6
      Football152.446.082.444.641.958.182.456.7
      Jumping16.690.195.917.34.813.462.381.3
      Liquor20.556.641.123.640.823.620.291.3
      Matrix2.01.012.02.021.034.012.036.0
      Skiing7.47.43.711.17.49.97.430.4
      Ironman10.83.04.813.310.814.57.213.9
      Jogging122.596.421.822.522.522.222.292.8
      Lemming28.060.566.216.946.043.087.290.8
      MotorRolling7.315.916.511.06.77.37.330.5
    • Table 3. Comparison of average center position error of tracking resultspixel

      View table

      Table 3. Comparison of average center position error of tracking resultspixel

      NameDLTTLDStruckASLADSSTSSTMILProposed algorithm
      Basketball12.0-118.382.6111.6105.991.910.8
      Bolt--398.8374.75.0409.9393.55.6
      David3107.4-106.587.888.4104.529.715.2
      Football110.4-5.512.220.515.75.68.9
      Jumping41.95.96.746.135.245.710.07.6
      Liquor153.3-91.0146.799.3146.7141.923.2
      Matrix171.1-194.865.259.754.755.033.5
      Skiing244.5-251.8266.6220.1269.9267.089.4
      Ironman211.4-127.6197.5105.7205.4193.474.3
      Jogging1113.06.762.0104.6112.0114.696.312.3
      Lemming128.9-37.8178.881.523.612.111.8
      MotorRolling170.7-145.6201.4289.6377.0161.0101.8
    Tools

    Get Citation

    Copy Citation Text

    Zefenfen Jin, Zhiqiang Hou, Wangsheng Yu, Xin Wang. Multiple Feature Fusion based on Covariance Matrix for Visual Tracking[J]. Acta Optica Sinica, 2017, 37(9): 0915005

    Download Citation

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

    Category: Machine Vision

    Received: Apr. 14, 2017

    Accepted: --

    Published Online: Sep. 7, 2018

    The Author Email: Zefenfen Jin (christine123456@163.com)

    DOI:10.3788/AOS201737.0915005

    Topics