Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Acta Optica Sinica, Volume. 37, Issue 8, 0815003(2017)

Robust Visual Tracking Based on Convolutional Neural Networks and Conformal Predictor

Lin Gao1、*, Junfeng Wang2, Yong Fan1, and Niannian Chen1
Author Affiliations
  • 1 School of Computer Science and Technology, Southwest University of Science and Technology, Mianyang, Sichuan 621010, China
  • 2 College of Computer Science, Sichuan University, Chengdu, Sichuan 610065, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (7)
    Equations (0)
    References (19)
    Cited By (4)
    Tools
    Paper Information
    Topics
    Figures & Tables(7)
    Flow chart of tracking algorithm

    Fig. 1. Flow chart of tracking algorithm

    Download full size
    Structure of CNN

    Fig. 2. Structure of CNN

    Download full size
    Tracking results of video sequences. (a) FaceOcc1; (b) Bolt; (c) Football; (d) CarDark

    Fig. 3. Tracking results of video sequences. (a) FaceOcc1; (b) Bolt; (c) Football; (d) CarDark

    Download full size
    Center position error of tracking results. (a) FaceOcc1; (b) Bolt; (c) Football; (d) CarDark

    Fig. 4. Center position error of tracking results. (a) FaceOcc1; (b) Bolt; (c) Football; (d) CarDark

    Download full size
    Coverage rate of tracking results. (a) FaceOcc1; (b) Bolt; (c) Football; (d) CarDark

    Fig. 5. Coverage rate of tracking results. (a) FaceOcc1; (b) Bolt; (c) Football; (d) CarDark

    Download full size
    One-pass evaluation. (a) Positional accuracy diagram; (b) coverage success rate diagram

    Fig. 6. One-pass evaluation. (a) Positional accuracy diagram; (b) coverage success rate diagram

    Download full size

    • Table 1. Average center position error and coverage rate

      View table

      Table 1. Average center position error and coverage rate

      SequenceVTSLOTStruckMILKCFProposed algorithm (no CP)Proposed algorithm
      Football13.27(0.51)9.23(0.54)13.33(0.53)12.55(0.58)14.60(0.55)6.73(0.66)5.36(0.68)
      FaceOcc127.42(0.57)34.22(0.40)24.50(0.63)34.86(0.54)15.98(0.75)14.29(0.73)20.43(0.70)
      CarDark23.45(0.45)37.43(0.26)3.42(0.75)45.69(0.15)6.05(0.61)10.54(0.39)3.04(0.74)
      Bolt197.53(0.02)13.18(0.45)360.19(0.01)387.04(0.01)6.37(0.68)33.25(0.22)8.65(0.65)
      Total average65.41(0.39)23.52(0.41)100.36(0.48)120.03(0.32)10.75(0.65)16.20(0.50)9.37(0.69)
    Tools

    Get Citation

    Copy Citation Text

    Lin Gao, Junfeng Wang, Yong Fan, Niannian Chen. Robust Visual Tracking Based on Convolutional Neural Networks and Conformal Predictor[J]. Acta Optica Sinica, 2017, 37(8): 0815003

    Download Citation

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

    Category: Machine Vision

    Received: Feb. 24, 2017

    Accepted: --

    Published Online: Sep. 7, 2018

    The Author Email: Gao Lin (gaolinscu@163.com)

    DOI:10.3788/AOS201737.0815003

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology