• Laser & Optoelectronics Progress
  • Vol. 57, Issue 18, 181006 (2020)
Jianjun Li, Yue Sun*, and Baohua Zhang
Author Affiliations
  • School of Information Engineering, Inner Mongolia University of Science and Technology, Baotou, Inner Mongolia 014010, China
  • show less
    DOI: 10.3788/LOP57.181006 Cite this Article Set citation alerts
    Jianjun Li, Yue Sun, Baohua Zhang. Interactive Behavior Recognition Based on Sparse Coding Feature Fusion[J]. Laser & Optoelectronics Progress, 2020, 57(18): 181006 Copy Citation Text show less
    Pixel gray value of 3×3 window
    Fig. 1. Pixel gray value of 3×3 window
    Neighborhood of circular symmetry. (a) (1,8); (b) (1.5,12); (c) (2,16)
    Fig. 2. Neighborhood of circular symmetry. (a) (1,8); (b) (1.5,12); (c) (2,16)
    Processing results of the two modes. (a) Gray image; (b) original LBP operator; (c) improved LBP operator
    Fig. 3. Processing results of the two modes. (a) Gray image; (b) original LBP operator; (c) improved LBP operator
    Neighborhood of pixel point
    Fig. 4. Neighborhood of pixel point
    Adjacent pixel
    Fig. 5. Adjacent pixel
    Edges extracted by different thresholds. (a) Depth image; (b) automatic threshold; (c) threshold range is [0.32,0.8]; (d) threshold range is [0.08,0.2]
    Fig. 6. Edges extracted by different thresholds. (a) Depth image; (b) automatic threshold; (c) threshold range is [0.32,0.8]; (d) threshold range is [0.08,0.2]
    Specific steps of the ScSPM model
    Fig. 7. Specific steps of the ScSPM model
    Flow chart of our algorithm
    Fig. 8. Flow chart of our algorithm
    Recognition results of the CAD-60 dataset
    Fig. 9. Recognition results of the CAD-60 dataset
    Recognition results of the MSR Action Pairs dataset
    Fig. 10. Recognition results of the MSR Action Pairs dataset
    Recognition results of the SBU Kinect interaction dataset
    Fig. 11. Recognition results of the SBU Kinect interaction dataset
    ReferenceRecognitionmethodRecognitionrate /%
    Ref. [16]Full tow-layer MEMM61.7
    Ref. [17]HMM82.3
    Ref. [18]CFt+RBF-SVM94.0
    OursS+ScSPM99.0
    Table 1. Recognition rates of different algorithms (CAD-60 dataset)
    ReferenceRecognitionmethodRecognitionrate /%
    Ref. [19]skeleton+LOPskeleton+LOP+pyramid61.782.2
    Ref. [20]DMM66.1
    Ref. [21]HON4D93.3
    OursS+ScSPM93.1
    Table 2. Recognition rates of different algorithms (MSR Action Pairs dataset)
    ReferenceRecognitionmethodRecognitionrate /%
    Ref. [22]joint features+CFDM89.4
    Ref. [23]SVM+LCNN92.8
    Ref. [24]BOW+HOG92.5
    Ref. [25]motion feature+shape feature(depth)98.4
    OursS+ScSPM95.4
    Table 3. Recognition rates of different algorithms (SBU Kinect interaction dataset)
    Jianjun Li, Yue Sun, Baohua Zhang. Interactive Behavior Recognition Based on Sparse Coding Feature Fusion[J]. Laser & Optoelectronics Progress, 2020, 57(18): 181006
    Download Citation