• Laser & Optoelectronics Progress
  • Vol. 60, Issue 12, 1210007 (2023)
Jing Yang and Long Ma*
Author Affiliations
  • School of Ordnance Science and Technology, Xi'an Technological University, Xi'an 710021, Shaanxi, China
  • show less
    DOI: 10.3788/LOP220929 Cite this Article Set citation alerts
    Jing Yang, Long Ma. Thermal Infrared Object Tracking Method Based on Positional Perception[J]. Laser & Optoelectronics Progress, 2023, 60(12): 1210007 Copy Citation Text show less
    Structure of the proposed network
    Fig. 1. Structure of the proposed network
    Schematic of dilated convolution
    Fig. 2. Schematic of dilated convolution
    Structure of positional perception module
    Fig. 3. Structure of positional perception module
    Thermal infrared object visualization
    Fig. 4. Thermal infrared object visualization
    Structure of channel attention module
    Fig. 5. Structure of channel attention module
    RPN module
    Fig. 6. RPN module
    Class activated thermal maps. (a) Without attention module; (b) with attention module
    Fig. 7. Class activated thermal maps. (a) Without attention module; (b) with attention module
    Some tracking results
    Fig. 8. Some tracking results
    Evaluation results. (a) Precision; (b) success rate
    Fig. 9. Evaluation results. (a) Precision; (b) success rate
    SequenceTotal framesChallenge
    aftertree313MC,OCC,SV,AI
    green491MC,SV,CM
    redcar218MC,SV
    woman89434CM,IV,OCC,AI,MC
    Table 1. Evaluation sequence
    MethodVOT-TIR2019GTOTSpeed /(frame·s-1
    AUCPrecisionAUCPrecision
    Proposed method0.5380.7530.6450.91430
    C-COT0.5460.7180.5990.8710.3
    ECO0.5350.7000.6000.8386
    SiamRPN0.4990.6720.6150.899160
    SiamFC0.4260.5940.5340.83958
    DSST0.3910.5330.3590.52424
    Table 2. Comparison of experimental results of 6 trackers on VOT-TIR2019 and GTOT datasets