Infrared and Laser Engineering, Volume. 53, Issue 11, 20240260(2024)

An RGBT progressive fusion visual tracking with time-domain updated templates

Yong GUO, Haiyun SHEN, Jianyu CHEN, and Jiemin YUAN
Author Affiliations
  • School of Electrical Information, Southwest Petroleum University, Chengdu 610500, China
  • show less
    Figures & Tables(12)
    The overall framework of SiamDPF
    The framework of MDAM
    The framework of CPFM
    The framework of CUTM
    The evaluation curve on GTOT dataset
    The evaluation curve on RGB234 dataset
    The challenge attribute comparison curve on RGB234 dataset
    Visualization results of in RGBT234
    • Table 1. The parameters structure of backbone network

      View table
      View in Article

      Table 1. The parameters structure of backbone network

      LayerKernelStrideTemplateImage
      127×127×3303×303×3
      Conv111×11259×59×96147×147×96
      Pool13×3229×29×9673×73×96
      Conv25×5125×25×25669×69×256
      Pool23×3212×12×25634×34×256
      Conv33×3110×10×38432×32×384
      MDAM--8×8×38430×30×384
      MDAM--6×6×25628×28×256
    • Table 2. The comparison of a series of algorithms results

      View table
      View in Article

      Table 2. The comparison of a series of algorithms results

      YearMethodFPSPRSR
      2021SiamCDA3787.7%73.2%
      2022DuSiamRT11676.6%62.8%
      2023SiamIVFN14791.5%79.3%
      2023SiamCAF10590.6%73.0%
      2023SiamTDR12788.5%71.4%
      SiamDPF (Ours)6891.6%73.5%
    • Table 3. The result of ablation experiments

      View table
      View in Article

      Table 3. The result of ablation experiments

      GroupMDAMCPFMCUTMGTOT
      PRSRFPS/frame·s–1
      1ΟΟΟ0.8830.719114
      2ΠΟΟ0.8920.72492
      3ΠΠΟ0.9090.73276
      4ΠΠΠ0.9160.73568
    • Table 4. The real-time analysis of network parameters

      View table
      View in Article

      Table 4. The real-time analysis of network parameters

      GroupMDAMCPFMCUTMParameters/×106FPS/frame·s–1
      1ΟΟΟ-114
      2ΠΟΟ12.5492
      3ΠΠΟ16.9776
      4ΠΠΠ18.7368
    Tools

    Get Citation

    Copy Citation Text

    Yong GUO, Haiyun SHEN, Jianyu CHEN, Jiemin YUAN. An RGBT progressive fusion visual tracking with time-domain updated templates[J]. Infrared and Laser Engineering, 2024, 53(11): 20240260

    Download Citation

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

    Category: 图像处理

    Received: Jun. 14, 2024

    Accepted: --

    Published Online: Dec. 13, 2024

    The Author Email:

    DOI:10.3788/IRLA20240260

    Topics