MAO Lin, CHEN Siyu, YANG Dawei. A Guided Method for Improving the Video Human Action Classification in Convolutional Neural Networks[J]. Geomatics and Information Science of Wuhan University, 2021, 46(8): 1241-1246. DOI: 10.13203/j.whugis20190101
Citation: MAO Lin, CHEN Siyu, YANG Dawei. A Guided Method for Improving the Video Human Action Classification in Convolutional Neural Networks[J]. Geomatics and Information Science of Wuhan University, 2021, 46(8): 1241-1246. DOI: 10.13203/j.whugis20190101
shu

A Guided Method for Improving the Video Human Action Classification in Convolutional Neural Networks

    Graphical Abstract
    • Abstract
  •   Objectives  In order to improve the ability of convolutional neural networks (CNNs) of understanding temporal dynamic information, this paper proposes a dominant layer optimization module.
      Methods  The new module uses the dominant layer to guide and optimize the update gradient of convolutional layer weights, and assist the difference estimation with the maximum mean difference algorithm of a reproducing Hilbert space.
      Results  In continuous training, the network can improve the learning ability of temporal dynamic information, and the dynamic information similarity between the features learned by convolutional layer and the input data is also increased.
      Conclusions  This module enhances the performance of the CNNs model on video human action classification and achieves improvements to the network.
    • FullText(HTML)
  • loading
    • References (21)
    • Related Articles

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return