Traditional pose resolution methods face the problem of insufficient accuracy when dealing with complex table tennis sports scenes. For this reason, this paper firstly extracts the foreground of table tennis movement based on the improved AlphaPose algorithm. Then a three-dimensional stacked convolutional neural network (3D-SCNN) is designed to extract the spatio-temporal features of the motion pose. A 3D convolutional block attention mechanism is proposed to enhance the important spatio-temporal features and output the final motion pose estimation results through fully connected layers. Finally, a similarity calculation method is used to calculate the cumulative distances of the key points in the standard and estimated movements, and scoring is performed, so as to realize the accurate analysis of the movement postures. The experimental outcome implies that the correct attitude estimation rate of 3D-SCNN is improved by at least 2.6%, which validates the effectiveness of the designed approach.
GuYYuCShaoS, et al.Effects of table tennis multi-ball training on dynamic posture control. PeerJ2019; 6: 1–19.
2.
OagazHSchounBChoiM-H. Real-time posture feedback for effective motor learning in table tennis in virtual reality. Int J Hum Comput Stud2022; 158: 10–19.
3.
YinJ. The method of table tennis players' posture recognition based on a genetic algorithm. Int J Biometrics2021; 13: 243–257.
4.
LiW. Analyzing the rotation trajectory in table tennis using deep learning. Soft Comput2023; 27: 12769–12785.
5.
Barczyk-PawelecKBańkoszZDerlichM. Body postures and asymmetries in frontal and transverse planes in the trunk area in table tennis players. Biol Sport2012; 29(2): 129–134.
6.
ChenYLiLLiX. Correlation analysis of structural characteristics of table tennis players' hitting movements and hitting effects based on data analysis. Entertainment Computing2024; 48: 16–29.
7.
LinYWuZZhangL. Optimization of table tennis players' technical movements based on genetic algorithm. Int J Multiphys2024; 18: 23–35.
8.
HeoGHaJ-E. Analysis of table tennis swing using action recognition. J Inst Control, Robot Syst2015; 21: 40–45.
9.
LikasAVlassisNVerbeekJJ. The global k-means clustering algorithm. Pattern Recogn2003; 36: 451–461.
10.
WangY. Rationality evaluation of batting action of table tennis players based on support vector machine model. J Sport Psychol2021; 30: 296–304.
11.
WuC-HWuT-CLinW-B. Exploration of applying pose estimation techniques in table tennis. Applied Sciences2023; 13: 1896.
12.
TabriziSSPashazadehSJavaniV. Comparative study of table tennis forehand strokes classification using deep learning and SVM. IEEE Sens J2020; 20: 13552–13561.
13.
RenW. A novel approach for automatic detection and identification of inappropriate postures and movements of table tennis players. Soft Comput2024; 28: 2245–2269.
14.
SunSZhengHLinZ. Optimization of table tennis swing action supported by the temporal convolutional network algorithm in deep learning. IEEE Access2024; 12: 178362–178374.
15.
PangY. Table tennis capture system based on image recognition and modeling. Sci Program2022; 20: 46–58.
16.
LiuQDingH. Application of table tennis ball trajectory and rotation-oriented prediction algorithm using artificial intelligence. Front Neurorobot2022; 16: 82–96.
ZhangJRenYLinL, et al.Table tennis motion recognition based on the bat trajectory using varying-length-input convolution neural networks. Sci Rep2024; 14: 35–49.
19.
BianJLiXWangT, et al.P2ANet: a large-scale benchmark for dense action detection from table tennis match broadcasting videos. ACM Trans Multimedia Comput Commun Appl2024; 20: 1–23.
20.
LuoHXiongCFangW, et al.Convolutional neural networks: computer vision-based workforce activity assessment in construction. Autom ConStruct2018; 94: 282–289.
21.
CongSZhouY. A review of convolutional neural network architectures and their optimizations. Artif Intell Rev2023; 56: 1905–1969.
22.
GamaFMarquesAGLeusG, et al.Convolutional neural network architectures for signals supported on graphs. IEEE Trans Signal Process2018; 67: 1034–1049.
23.
ChaobankohNJumphooTUthansakulM, et al.Lower-limb motion-based ankle-foot movement classification using 2D-CNN. Comput Mater Contin2022; 73: 1269–1282.
24.
FrintropSRomeEChristensenHI. Computational visual attention systems and their cognitive foundations: a survey. ACM Trans Appl Percept2010; 7: 1–39.
25.
WangYWangWLiY, et al.An attention mechanism module with spatial perception and channel information interaction. Complex Intell Systems2024; 10: 5427–5444.
26.
ChenZTianSYuL, et al.An object detection network based on YOLOv4 and improved spatial attention mechanism. J Intell Fuzzy Syst2022; 42: 2359–2368.
27.
ZhangJ. A novel defect segmentation model via integrating enhanced U-net with convolutional block attention mechanisms. Appl Comp Eng2024; 103: 178–190.
LiSHuoH. Frame deletion detection based on optical flow orientation variation. IEEE Access2021; 9: 37196–37209.
30.
ChengYHWangJ. A motion image detection method based on the inter-frame difference method. Appl Mech Mater2014; 490: 1283–1286.
31.
FangH-SLiJTangH, et al.Alphapose: whole-body regional multi-person pose estimation and tracking in real-time. IEEE Trans Pattern Anal Mach Intell2022; 45: 7157–7173.
32.
MehdizadehSNabaviHSaboA, et al.The toronto older adults gait archive: video and 3d inertial motion capture data of older adults’ walking. Sci Data2022; 9: 398.
33.
OlorunsholaOEIrhebhudeMEEvwiekpaefeAE. A comparative study of YOLOv5 and YOLOv7 object detection algorithms. J Comp Soci Infor2023; 2: 1–12.
34.
WangLWangZZhouL. Application of a two-dimensional regression network algorithm model based on local constraints in human motion recognition. IEEE Access2024; 12: 32253–32265.
35.
QiaoF. Application of deep learning in automatic detection of technical and tactical indicators of table tennis. PLoS One2021; 16: 1–16.
36.
ZhangXZhouZHanY, et al.Deep learning-based real-time 3D human pose estimation. Eng Appl Artif Intell2023; 119: 58–73.
37.
WuTPanCPanY, et al.Joint angle estimation error analysis and 3-D positioning algorithm design for mmWave positioning system. IEEE Internet Things J2023; 11: 2181–2197.
