Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 56, Issue 12, 121004(2019)

Video Classification Based on Three-Dimensional Squeeze Excitation Module

Ningxiao Li, Guodong Wang*, Yanjie Wang, Shiyu Hu, and Liangliang Wang
Author Affiliations
  • College of Computer Science & Technology, Qingdao University, Qingdao, Shandong 266071, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (9)
    Equations (0)
    References (25)
    Tools
    Paper Information
    Topics
    Figures & Tables(9)
    Unit structure of module. (a) 2D ResNet unit; (b) 3D ResNet unit; (c) 3D SE_FC ResNet unit

    Fig. 1. Unit structure of module. (a) 2D ResNet unit; (b) 3D ResNet unit; (c) 3D SE_FC ResNet unit

    Download full size
    Flow chart of feature extraction

    Fig. 2. Flow chart of feature extraction

    Download full size
    Loss at first stage of UCF101 training

    Fig. 3. Loss at first stage of UCF101 training

    Download full size
    Loss at second stage of UCF101 training

    Fig. 4. Loss at second stage of UCF101 training

    Download full size
    Loss of HMDB51 training

    Fig. 5. Loss of HMDB51 training

    Download full size

    • Table 1. Average validation accuracy comparison on UCF101 dataset

      View table

      Table 1. Average validation accuracy comparison on UCF101 dataset

      MethodPretrainingdatasetAverage validationaccuracy /%
      Top1Top5
      3D ResNet5043.867.3
      3D ResNet10143.468.8
      3D SE_FC ResNet50,T=1No42.566.1
      3D SE_FC ResNet50,T=444.969.6
      3D SE_FC ResNet50,T=845.568.2

    • Table 2. Average validation accuracy comparison on HMDB51 dataset

      View table

      Table 2. Average validation accuracy comparison on HMDB51 dataset

      MethodPretrainingdatasetAverage validationaccuracy /%
      Top1Top5
      3D ResNet5015.643.0
      3D ResNet10115.241.0
      3D SE_FC ResNet50,T=1UCF10116.143.6
      3D SE_FC ResNet50,T=418.746.2
      3D SE_FC ResNet50,T=817.744.6

    • Table 3. Average time comparison for each video classification

      View table

      Table 3. Average time comparison for each video classification

      MethodTime /ms
      3D ResNet50104
      3D ResNet101111
      3D SE_FC ResNet50,T=1112
      3D SE_FC ResNet50,T=4110
      3D SE_FC ResNet50,T=8114

    • Table 4. Test accuracy comparison on HMDB51 and UCF101 datasets

      View table

      Table 4. Test accuracy comparison on HMDB51 and UCF101 datasets

      MethodPretrainingdatasetTest accuracy /%
      HMDB51UCF101
      3D ResNet18Kinetics56.484.4
      3D ResNet3459.187.7
      3D ResNet5061.089.3
      3D ResNet10161.788.9
      3D ResNet20063.589.6
      3D DenseNet-12159.687.6
      Method in Ref. [21]-59.488.0
      Method in Ref.[22]--88.6
      Method in Ref. [23]--85.9
      Method in Ref. [24]--88.6
      Method in Ref.[25]sports 1M-82.3
      3D SE_FC ResNet50,T=1(ours)Kinetics61.389.0
      3D SE_FC ResNet50,T=4(ours)59.690.1
      3D SE_FC ResNet50,T=8(ours)59.089.5
    Tools

    Get Citation

    Copy Citation Text

    Ningxiao Li, Guodong Wang, Yanjie Wang, Shiyu Hu, Liangliang Wang. Video Classification Based on Three-Dimensional Squeeze Excitation Module[J]. Laser & Optoelectronics Progress, 2019, 56(12): 121004

    Download Citation

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

    Category: Image Processing

    Received: Nov. 29, 2018

    Accepted: Jan. 11, 2019

    Published Online: Jun. 13, 2019

    The Author Email: Wang Guodong (doctorwgd@gmail.com)

    DOI:10.3788/LOP56.121004

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology