Laser & Optoelectronics Progress, Volume. 57, Issue 24, 241506(2020)

Improved Human Action Recognition Algorithm Based on Two-Stream Faster Region Convolutional Neural Network

Ruyi Guo, Jie Jin*, Gaohua Liu, Kaiyan Liu, and Shiqi Jiang
Author Affiliations
  • School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
  • show less
    Figures & Tables(6)
    Overall structure of our network
    Structure of SE-VGG16
    Structure of Improved SE-VGG16
    Detection results of our algorithm. (a) Walking with dog; (b) diving; (c) biking;(d) floor gymnastics
    • Table 1. mAP of different algorithms on UCF101 data set unit: %

      View table

      Table 1. mAP of different algorithms on UCF101 data set unit: %

      MethodmAP
      IDT86.4
      C3D85.2
      TDD90.3
      Two-stream88.0
      ST-ResNet93.4
      Ours92.8
    • Table 2. Influence of different modules on our algorithm unit: %

      View table

      Table 2. Influence of different modules on our algorithm unit: %

      MethodVGG16SE-VGG16Improved SE-VGG16IOU lossCIOU lossmAP
      Two-stream Faster RCNN90.5
      Two-stream Faster RCNN92.1
      Two-stream Faster RCNN92.5
      Two-stream Faster RCNN92.8
    Tools

    Get Citation

    Copy Citation Text

    Ruyi Guo, Jie Jin, Gaohua Liu, Kaiyan Liu, Shiqi Jiang. Improved Human Action Recognition Algorithm Based on Two-Stream Faster Region Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2020, 57(24): 241506

    Download Citation

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

    Category: Machine Vision

    Received: Apr. 29, 2020

    Accepted: Jun. 17, 2020

    Published Online: Dec. 1, 2020

    The Author Email: Jin Jie (jinjie@tju.edu.cn)

    DOI:10.3788/LOP57.241506

    Topics