Chinese Journal of Liquid Crystals and Displays, Volume. 40, Issue 4, 598(2025)

Pose estimation network based on attention feature fusion of multimodal data

Yuntao ZHAO and Xinhui DENG*
Author Affiliations
  • College of Information Science and Engineering,Wuhan University of Science and Technology.,Wuhan 430081,China
  • show less
    Figures & Tables(10)
    Attention feature fusion pose estimation network
    SE attention module architecture diagram
    Original ResNet(a)and SE-ResNet variant network(b)
    iAFF module
    MS-CAM module(a)and AFF module(b)
    Pose evaluation visualization on YCB-video dataset
    Pose evaluation visualization on LineMOD dataset
    • Table 1. Comparison of methods on YCB-video dataset

      View table
      View in Article

      Table 1. Comparison of methods on YCB-video dataset

      模型类别平均值
      Cracker boxTomato soup canMustard bottlePudding boxAUC
      AUC< 2 cmAUC< 2 cmAUC< 2 cmAUC< 2 cm
      Pointfusion80.562.691.996.988.584.087.596.786.1
      Posecnn92.783.494.596.990.496.897.996.992.7
      Densefusion90.898.492.995.791.297.888.397.289.8
      Ours96.899.693.895.797.810095.999.595.1
      Potted meat canPitcher baseBowlMug<2 cm
      AUC< 2 cmAUC< 2 cmAUC< 2 cmAUC< 2 cm
      Pointfusion86.488.585.579.875.724.192.499.879.1
      Posecnn92.793.697.896.981.081.895.099.893.2
      Densefusion87.391.286.995.690.999.790.394.796.3
      Ours90.392.298.110091.098.797.610098.2
    • Table 2. Comparison of ADD metric on LineMOD dataset %

      View table
      View in Article

      Table 2. Comparison of ADD metric on LineMOD dataset %

      模型类别
      ApeBen.CamCan
      Densefusion92.694.396.594.4
      Ours94.496.296.395.9
      CatDrillDuckEgg.
      Densefusion96.687.793.399.9
      Ours96.995.095.399.9
      平均值Dense.95.3Ours96.7
    • Table 3. Comparison of OP metric on LineMOD dataset %

      View table
      View in Article

      Table 3. Comparison of OP metric on LineMOD dataset %

      模型类别
      ApeBen.CamCan
      Densefusion92.694.396.594.4
      Ours94.496.296.395.9
      CatDrillDuckEgg.
      Densefusion96.687.793.399.9
      Ours96.995.095.399.9
      平均值Dense.80.5Ours87.4
    Tools

    Get Citation

    Copy Citation Text

    Yuntao ZHAO, Xinhui DENG. Pose estimation network based on attention feature fusion of multimodal data[J]. Chinese Journal of Liquid Crystals and Displays, 2025, 40(4): 598

    Download Citation

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

    Category:

    Received: Jul. 28, 2024

    Accepted: --

    Published Online: May. 21, 2025

    The Author Email: Xinhui DENG (2211241803@qq.com)

    DOI:10.37188/CJLCD.2024-0218

    Topics