• Laser & Optoelectronics Progress
  • Vol. 57, Issue 14, 141504 (2020)
Yuting Su, Mengmeng Wang, Jing Liu*, Yunpeng Fei, and Xu He
Author Affiliations
  • School of Electrical and Information Engineering, Tianjin University, Tianjin 300072 China
  • show less
    DOI: 10.3788/LOP57.141504 Cite this Article Set citation alerts
    Yuting Su, Mengmeng Wang, Jing Liu, Yunpeng Fei, Xu He. Micro-Expression Recognition Algorithm Based on Multiple Motive Feature Fusion[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141504 Copy Citation Text show less
    Overall network framework
    Fig. 1. Overall network framework
    Diagram of LMF detection process. (a) Example of 68 monitoring points in face; (b) LMF visualization
    Fig. 2. Diagram of LMF detection process. (a) Example of 68 monitoring points in face; (b) LMF visualization
    Diagram of optical flow and optical strain. (a) Start frame of micro-expression; (b) peak frame of micro-expression;(c) optical flow visualization image; (d) optical strain visualization image
    Fig. 3. Diagram of optical flow and optical strain. (a) Start frame of micro-expression; (b) peak frame of micro-expression;(c) optical flow visualization image; (d) optical strain visualization image
    Input dataAccuracy /%
    Optical strain47.98
    LMF45.70
    Optical flow41.50
    Original data36.65
    Table 1. Performance comparison of different data in CNN module
    ExperimentNo.OpticalstrainLMFOpticalflowAccuracy /%
    152.45
    251.38
    350.54
    458.53
    Table 2. Result comparison of single feature map and multiple feature fusion map
    AlgorithmAccuracy /%UAR /%F1-score /%
    LBP-TOP*45.9530.9429.41
    LBP-SIPr46.5644.80
    FDMo45.9340.53
    ELRCN-TE*50.0039.3743.4
    MRWo46.1543.07
    STCLQPo58.3958.36
    MDMOr44.2544.16
    Bi-WOOFo57.8961.25
    Proposed algorithm58.5348.0752.82
    Table 3. Comparison of indicators of different algorithms
    Yuting Su, Mengmeng Wang, Jing Liu, Yunpeng Fei, Xu He. Micro-Expression Recognition Algorithm Based on Multiple Motive Feature Fusion[J]. Laser & Optoelectronics Progress, 2020, 57(14): 141504
    Download Citation