# Table Tennis Track Detection Based on Temporal Feature Multiplexing Network

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Related Work

## 3. Materials and Methods

#### 3.1. Design of the Main Structure of the Network

#### 3.2. Design of Feature Information Return Module

#### 3.3. Design of the Location Information Return Module

## 4. Results Analysis and Discussion

#### 4.1. Introduction to Dataset Production and Experimental Environment

#### 4.2. Ablation Experiments

#### 4.3. Cross-Sectional Comparison Experiments

#### 4.4. Discussion

## 5. Conclusions and Future Work

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- He, C.; Zhang, X.; Gui, Y.; Liu, Y.; Zhang, W. Mathematical modeling and simulation of table tennis trajectory based on digital video image processing. Adv. Math. Phys.
**2021**, 2021, 7045445. [Google Scholar] [CrossRef] - Zhou, X. Explanation and verification of the rules of attack in table tennis tactics. BMC Sports Sci. Med. Rehabil.
**2022**, 14, 1–8. [Google Scholar] [CrossRef] - Zhao, L.; Pan, D. Video analysis of belt and road sports events based on wireless network and artificial intelligence technology. Wirel. Commun. Mob. Comput.
**2022**, 2022, 8278045. [Google Scholar] [CrossRef] - Cao, Z.; Liao, T.; Song, W.; Chen, Z.; Li, C. Detecting the shuttlecock for a badminton robot: A YOLO based approach. Expert Syst. Appl.
**2021**, 164, 113833. [Google Scholar] [CrossRef] - Fu, X.B.; Yue, S.L.; Pan, D.Y. Camera-based basketball scoring detection using convolutional neural network. Int. J. Autom. Comput.
**2020**, 18, 266–276. [Google Scholar] [CrossRef] - Zhang, X.F.; Dai, L.L.; Zhou, J.Q.; Chu, Y.Z.; Bin, J.I. Quick detection and real-time tracking for table tennis ball based on multi-color models. Comput. Syst. Appl.
**2018**, 27, 232–237. (In Chinese) [Google Scholar] - Wang, Q.; Zhang, K.; Wang, D. The trajectory prediction and analysis of spinning ball for a table tennis robot application. In Proceedings of the 4th Annual IEEE International Conference on Cyber Technology in Automation, Control and Intelligent (IEEE-CYBER), Hong Kong, China, 4–7 June 2014; pp. 496–501. [Google Scholar]
- Qiao, F. Application of deep learning in automatic detection of technical and tactical indicators of table tennis. PLoS ONE
**2021**, 16, e0245259. [Google Scholar] [CrossRef] - Lin, H.I.; Yu, Z.; Huang, Y.C. Ball tracking and trajectory prediction for table-tennis robots. Sensors
**2020**, 20, 333. [Google Scholar] [CrossRef] - Zhang, J. Automatic detection method of technical and tactical indicators for table tennis based on trajectory prediction using compensation fuzzy neural network. Comput. Intell. Neurosci.
**2021**, 2021, 3155357. [Google Scholar] [CrossRef] - Zhao, H.; Hao, F. Target tracking algorithm for table tennis using machine vision. J. Healthc. Eng.
**2021**, 2021, 9961978. [Google Scholar] [CrossRef] - Oagaz, H.; Schoun, B.; Choi, M.H. Real-time posture feedback for effective motor learning in table tennis in virtual reality. Int. J. Hum. Comput. Stud.
**2022**, 158, 102731. [Google Scholar] [CrossRef] - Yang, H.H.; Huang, K.C.; Chen, W.T. Laffnet: A lightweight adaptive feature fusion network for underwater image enhancement. In Proceedings of the 2021 IEEE International Conference on Robotics and Automation (ICRA), Xi’an, China, 30 May–5 June 2021; pp. 774–785. [Google Scholar]
- Jiao, C.; Chen, C.; Gou, S.; Wang, X.; Jiao, L. L₁ Sparsity-Regularized Attention Multiple-Instance Network for Hyperspectral Target Detection. IEEE Trans. Cybern.
**2021**, 53, 124–137. [Google Scholar] [CrossRef] - Junos, M.H.; Khairuddin, A.; Thannirmalai, S.; Dahari, M. An optimized yolo-based object detection model for crop harvesting system. IET Image Process.
**2021**, 15, 2112–2125. [Google Scholar] [CrossRef] - Himeur, C.E.; Lejemble, T.; Pellegrini, T.; Paulin, M.; Barthe, L.; Mellado, N. PCEDNet: A Lightweight Neural Network for Fast and Interactive Edge Detection in 3D Point Clouds. ACM Trans. Graph.
**2021**, 41, 1–21. [Google Scholar] [CrossRef] - Qin, C.; Wu, R.; Huang, G.; Tao, J.; Liu, C. A novel LSTM-autoencoder and enhanced transformer-based detection method for shield machine cutterhead clogging. Sci. China Technol. Sci.
**2023**, 2023, 1–16. [Google Scholar] [CrossRef] - Gao, C.; Yan, J.; Peng, X.; Liu, H. Signal structure information-based target detection with a fully convolutional network. Inf. Sci.
**2021**, 576, 345–354. [Google Scholar] [CrossRef] - Wu, X.; Wei, X.; Zhang, L.; Xu, H.; He, W.; Feng, Z.; Han, X.; Pei, B. Vmfnet: Visual-microwave dual-modality real-time target detection model for detecting damage to curved radar absorbing materials. Opt. Express
**2021**, 29, 23182–23201. [Google Scholar] [CrossRef] - Fang, L.; Sun, M. Motion recognition technology of badminton players in sports video images. Future Gener. Comput. Syst.
**2021**, 124, 381–389. [Google Scholar] [CrossRef] - Pan, Z.; Li, C. Robust basketball sports recognition by leveraging motion block estimation. Signal Process. Image Commun.
**2020**, 83, 115784. [Google Scholar] [CrossRef] - Sun, C.; Ma, D. Svm-based global vision system of sports competition and action recognition. Int. J. Intell. Syst.
**2021**, 40, 2265–2276. [Google Scholar] [CrossRef] - Mei, Z.; Wang, Y. Research on moving target detection and tracking technology in sports video based on sift algorithm. Adv. Multimed.
**2022**, 2022, 2743696. [Google Scholar] [CrossRef] - Ong, P.; Chong, T.K.; Ong, K.M.; Low, E.S. Tracking of moving athlete from video sequences using flower pollination algorithm. Vis. Comput.
**2021**, 38, 939–962. [Google Scholar] [CrossRef] - Naik, B.T.; Hashmi, M.F.; Bokde, N.D. A comprehensive review of computer vision in sports: Open issues, future trends and research directions. Appl. Sci.
**2022**, 12, 4429. [Google Scholar] [CrossRef] - Liu, L. Objects detection toward complicated high remote basketball sports by leveraging deep cnn architecture. Future Gener. Comput. Syst.
**2021**, 119, 31–36. [Google Scholar] [CrossRef] - Sun, Y.; Li, Y. A Deep Learning Method for Intelligent Analysis of Sports Training Postures. Comput. Intell. Neurosci.
**2022**, 2022, 2442606. [Google Scholar] [CrossRef] - Duan, C. Deep learning-based multitarget motion shadow rejection and accurate tracking for sports video. Complexity
**2021**, 2021, 5973531. [Google Scholar] [CrossRef] - Huang, Z. Accurate Recognition Method of Continuous Sports Action Based on Deep Learning Algorithm. Wirel. Commun. Mob. Comput.
**2022**, 2022, 3407935. [Google Scholar] - Baclig, M.M.; Ergezinger, N.; Mei, Q.; Gül, M.; Adeeb, S.; Westover, L. A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash. Appl. Sci.
**2020**, 10, 8793. [Google Scholar] [CrossRef] - Shah, C.; Du, Q.; Xu, Y. Enhanced TabNet: Attentive interpretable tabular learning for hyperspectral image classification. Remote Sens.
**2022**, 14, 716. [Google Scholar] [CrossRef] - Qin, C.; Jin, Y.; Zhang, Z.; Yu, H.; Tao, J.; Sun, H.; Liu, C. Anti-noise diesel engine misfire diagnosis using a multi-scale CNN-LSTM neural network with denoising module. CAAI Trans. Intell. Technol.
**2023**, 2023, 1–24. [Google Scholar] [CrossRef] - Hou, Y.; Yang, Q.; Li, L.; Shi, G. Detection and Recognition Algorithm of Arbitrary-Oriented Oil Replenishment Target in Remote Sensing Image. Sensors
**2023**, 23, 767. [Google Scholar] [CrossRef] - Munteanu, D.; Moina, D.; Zamfir, C.G.; Petrea, Ș.M.; Cristea, D.S.; Munteanu, N. Sea Mine Detection Framework Using YOLO, SSD and EfficientDet Deep Learning Models. Sensors
**2022**, 22, 9536. [Google Scholar] [CrossRef] - Qin, C.; Huang, G.; Yu, H.; Wu, R.; Tao, J.; Liu, C. Geological information prediction for shield machine using an enhanced multi-head self-attention convolution neural network with two-stage feature extraction. Geosci. Front.
**2023**, 14, 101519. [Google Scholar] [CrossRef] - Gu, X.; See, K.W.; Li, P.; Shan, K.; Wang, Y.; Zhao, L.; Lim, K.C.; Zhang, N. A novel state-of-health estimation for the lithium-ion battery using a convolutional neural network and transformer model. Energy
**2023**, 262, 125501. [Google Scholar] [CrossRef] - Zhang, X.; Xuan, J.; Yao, C.; Gao, Q.; Wang, L.; Jin, X.; Li, S. A deep learning approach for orphan gene identification in moso bamboo (Phyllostachys edulis) based on the CNN+ Transformer model. BMC Bioinform.
**2022**, 23, 1–19. [Google Scholar] [CrossRef] - Liu, H.; Liu, Z.; Jia, W.; Zhang, D.; Wang, Q.; Tan, J. Tool wear estimation using a cnn-transformer model with semi-supervised learning. Meas. Sci. Technol.
**2021**, 32, 125010. [Google Scholar] [CrossRef] - Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, L.; Polosukhin, I. Attention is all you need. arXiv
**2017**, arXiv:1706.03762v5. [Google Scholar] - Mehboob, F.; Rauf, A.; Jiang, R.; Saudagar, A.K.J.; Malik, K.M.; Khan, M.B.; Hasnat, M.H.A.; AlTameem, A.; AlKhathami, M. Towards robust diagnosis of COVID-19 using vision self-attention transformer. Sci. Rep.
**2022**, 12, 8922. [Google Scholar] [CrossRef] - Chen, H.; Li, C.; Wang, G.; Li, X.; Mamunur, R.M.; Sun, H.; Hu, W.; Li, Y.; Liu, W.; Sun, C. Gashis-transformer: A multi-scale visual transformer approach for gastric histopathological image detection. Pattern Recognit.
**2022**, 130, 108827. [Google Scholar] [CrossRef] - Zhou, Y.; Ji, A.; Zhang, L. Sewer defect detection from 3D point clouds using a transformer-based deep learning model. Autom. Constr.
**2022**, 136, 104163. [Google Scholar] [CrossRef] - Wang, L.; Zhu, H.; He, Z.; Jia, Y.; Du, J. Adjacent slices feature transformer network for single anisotropic 3D brain MRI image super-resolution. Biomed. Signal Process. Control
**2022**, 72, 103339. [Google Scholar] [CrossRef] - Dai, S.; Zhu, Y.; Jiang, X.; Yu, F.; Lin, J.; Yang, D. TD-Net: Trans-Deformer network for automatic pancreas segmentation. Neurocomputing
**2023**, 517, 279–293. [Google Scholar] [CrossRef] - Yang, J.; Tu, J.; Zhang, X.; Yu, S.; Zheng, X. TSE DeepLab: An efficient visual transformer for medical image segmentation. Biomed. Signal Process. Control
**2023**, 80, 104376. [Google Scholar] [CrossRef] - Zhang, N.; Yu, L.; Zhang, D.; Wu, W.; Tian, S.; Kang, X. Apt-net: Adaptive encoding and parallel decoding transformer for medical image segmentation. Comput. Biol. Med.
**2022**, 151, 106292. [Google Scholar] [CrossRef] - Wang, S.; Wang, H.; She, S.; Zhang, Y.; Qiu, Q.; Xiao, Z. Swin-T-NFC CRFs: An encoder–decoder neural model for high-precision UAV positioning via point cloud super resolution and image semantic segmentation. Comput. Commun.
**2023**, 197, 52–60. [Google Scholar] [CrossRef] - Wang, F.; Tian, S.; Yu, L.; Liu, J.; Wang, J.; Li, K.; Wang, Y. TEDT: Transformer-Based Encoding–Decoding Translation Network for Multimodal Sentiment Analysis. Cognit. Comput.
**2022**, 2022, 1–15. [Google Scholar] [CrossRef] - Lin, C.; Jiang, Y.; Cai, J.; Qu, L.; Haffari, G.; Yuan, Z. Multimodal transformer with variable-length memory for vision-and-language navigation. In Proceedings of the European Conference on Computer Vision (ECCV), Tel Aviv, Israel, 23–27 October 2022; pp. 380–397. [Google Scholar]
- Ju, B.; Wang, Q.; Fang, L.; Liu, F.; Li, G.; Liu, Y. Single-layer piezoelectric transformers with a unique design of polarization topologies. Sens. Actuators A Phys.
**2021**, 332, 113193. [Google Scholar] [CrossRef] - Tummala, S.; Kadry, S.; Bukhari, S.A.C.; Rauf, H.T. Classification of brain tumor from magnetic resonance imaging using vision transformers ensembling. Curr. Oncol.
**2022**, 29, 7498–7511. [Google Scholar] [CrossRef] - Xie, T.; Zhang, Z.; Tian, J.; Ma, L. Focal DETR: Target-Aware Token Design for Transformer-Based Object Detection. Sensors
**2022**, 22, 8686. [Google Scholar] [CrossRef] - Zeng, K.; Ma, Q.; Wu, J.; Xiang, S.; Shen, T.; Zhang, L. Nlfftnet: A non-local feature fusion transformer network for multi-scale object detection. Neurocomputing
**2022**, 493, 15–27. [Google Scholar] [CrossRef] - Zheng, Y.; Sun, P.; Zhou, Z.; Xu, W.; Ren, Q. Adt-det: Adaptive dynamic refined single-stage transformer detector for arbitrary-oriented object detection in satellite optical imagery. Remote Sens.
**2021**, 13, 2623. [Google Scholar] [CrossRef] - Hendria, W.F.; Phan, Q.T.; Adzaka, F.; Jeong, C. Combining transformer and cnn for object detection in uav imagery. ICT Express
**2021**, 2021, 1–6. [Google Scholar] [CrossRef] - Wu, R.; Liu, C.; Han, T.; Yao, J.; Jiang, D. A planetary gearbox fault diagnosis method based on time-series imaging feature fusion and a transformer model. Meas. Sci. Technol.
**2022**, 34, 024006. [Google Scholar] [CrossRef] - Chen, D.; Wang, D.; Hu, H. Transformer with sparse self-attention mechanism for image captioning. Electron. Lett.
**2020**, 56, 764–766. [Google Scholar] - Vaidya, B.; Paunwala, C. Hardware efficient modified cnn architecture for traffic sign detection and recognition. Int. J. Image. Graph.
**2021**, 22, 2250017. [Google Scholar] [CrossRef] - Zhao, Z.L.; Huang, D.; Ma, X.X. Toast:Automated testing of object transformers in dynamic software updates. J. Comput. Sci. Technol.
**2022**, 37, 50–66. [Google Scholar] [CrossRef] - Li, Z.; Chen, G.; Zhang, T. A CNN-transformer hybrid approach for crop classification using multitemporal multisensor images. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2020**, 13, 847–858. [Google Scholar] [CrossRef] - Fan, X.; Feng, X.; Dong, Y.; Hou, H. COVID-19 ct image recognition algorithm based on transformer and cnn. Displays
**2022**, 72, 102150. [Google Scholar] [CrossRef] - Schmid, F.; Koutini, K.; Widmer, G. Efficient large-scale audio tagging via transformer-to-cnn knowledge distillation. arXiv
**2022**, arXiv:2211.04772. [Google Scholar] - Zhang, Y.; Zhang, S.; Li, Y.; Zhang, Y. Single-and cross-modality near duplicate image pairs detection via spatial transformer comparing cnn. Sensors
**2021**, 21, 255. [Google Scholar] [CrossRef] - Ormerod, M.; Del Rincón, J.M.; Devereux, B. Predicting semantic similarity between clinical sentence pairs using transformer models: Evaluation and representational analysis. JMIR Med. Inform.
**2021**, 9, e23099. [Google Scholar] [CrossRef] [PubMed] - Zhou, X.; Wu, X.; Ding, P.; Li, X.; He, N.; Zhang, G.; Zhang, X. Research on transformer partial discharge uhf pattern recognition based on cnn-lstm. Energies
**2019**, 13, 21. [Google Scholar] [CrossRef] - Xu, X.; Feng, Z.; Cao, C.; Li, M.; Wu, J.; Wu, Z.; Shang, Y.; Ye, S. An improved swin transformer-based model for remote sensing object detection and instance segmentation. Remote Sens.
**2021**, 13, 4779. [Google Scholar] [CrossRef] - Zhang, J.; Li, C.; Liu, G.; Min, M.; Wang, C.; Li, J.; Wang, Y.; Yan, H.; Zuo, Z.; Huang, W. A CNN-transformer hybrid approach for decoding visual neural activity into text. Comput. Methods Programs Biomed.
**2021**, 214, 106586. [Google Scholar] [CrossRef] - Shetty, S.; Mahale, A. MS-CheXNet: An Explainable and Lightweight Multi-Scale Dilated Network with Depthwise Separable Convolution for Prediction of Pulmonary Abnormalities in Chest Radiographs. Mathematics
**2022**, 10, 3646. [Google Scholar] [CrossRef] - Yu, H.; Qin, C.; Tao, J.; Liu, C.; Liu, Q. A multi-channel decoupled deep neural network for tunnel boring machine torque and thrust prediction. Tunn. Undergr. Sp. Tech.
**2023**, 133, 104949. [Google Scholar] [CrossRef] - Hassan, E. Scene text detection using attention with depthwise separable convolutions. Appl. Sci.
**2022**, 12, 6425. [Google Scholar] [CrossRef] - Bernardo, L.S.; Damaševičius, R.; Ling, S.H.; de Albuquerque, V.H.C.; Tavares, J.M.R. Modified squeezenet architecture for parkinson’s disease detection based on keypress data. Biomedicines
**2022**, 10, 2746. [Google Scholar] [CrossRef] - Tsivgoulis, M.; Papastergiou, T.; Megalooikonomou, V. An improved squeezenet model for the diagnosis of lung cancer in ct scans. Mach. Learn. Appl.
**2022**, 10, 100399. [Google Scholar] [CrossRef] - Yu, H.; Sun, H.; Tao, J.; Qin, C.; Xiao, D.; Jin, Y.; Liu, C. A multi-stage data augmentation and AD-ResNet-based method for EPB utilization factor prediction. Automat. Constr.
**2023**, 147, 104734. [Google Scholar] [CrossRef] - de la Fuente Castillo, V.; Díaz-Álvarez, A.; Manso-Callejo, M.Á.; Serradilla Garcia, F. Grammar guided genetic programming for network architecture search and road detection on aerial orthophotography. Appl. Sci.
**2020**, 10, 3953. [Google Scholar] [CrossRef] - Bianco, S.; Buzzelli, M.; Ciocca, G.; Schettini, R. Neural architecture search for image saliency fusion. Inf. Fusion
**2020**, 57, 89–101. [Google Scholar] [CrossRef] - Qin, C.; Xiao, D.; Tao, J.; Yu, H.; Jin, Y.; Sun, Y.; Liu, C. Concentrated velocity synchronous linear chirplet transform with application to robotic drilling chatter monitoring. Measurement
**2022**, 194, 111090. [Google Scholar] [CrossRef] - Ryoo, M.S.; Piergiovanni, A.J.; Arnab, A.; Dehghani, M.; Angelova, A. TokenLearner: What Can 8 Learned Tokens Do for Images and Videos? arXiv
**2021**, arXiv:2106.11297. [Google Scholar] - Jia, Q.; Shu, H. Bitr-unet: A cnn-transformer combined network for mri brain tumor segmentation. In Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Singapore, 18–22 September 2022; pp. 3–14. [Google Scholar]
- Yoo, J.; Kim, T.; Lee, S.; Kim, S.H.; Lee, H.; Kim, T.H. Rich cnn-transformer feature aggregation networks for super-resolution. arXiv
**2022**, arXiv:2203.07682. [Google Scholar] - Ma, J.; Tang, L.; Fan, F.; Huang, J.; Mei, X.; Ma, Y. SwinFusion: Cross-domain long-range learning for general image fusion via swin transformer. IEEE/CAA J. Automatic.
**2022**, 9, 1200–1217. [Google Scholar] [CrossRef] - Chen, H.; Yin, Z.; Zhang, P.; Liu, P. Sleepzznet: Sleep stage classification using single-channel eeg based on cnn and transformer. Int. J. Psychophysiol.
**2021**, 168, S168. [Google Scholar] [CrossRef] - Dobko, M.; Kolinko, D.I.; Viniavskyi, O.; Yelisieiev, Y. Combining CNNs with transformer for multimodal 3D MRI brain tumor segmentation with self-supervised pretraining. In Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Singapore, 18–22 September 2022; pp. 232–241. [Google Scholar]
- Gao, G.; Wang, Z.; Li, J.; Li, W.; Yu, Y.; Zeng, T. Lightweight bimodal network for single-image super-resolution via symmetric cnn and recursive transformer. arXiv
**2022**, arXiv:2204.13286. [Google Scholar] - Luo, X.; Hu, M.; Song, T.; Wang, G.; Zhang, S. Semi-supervised medical image segmentation via cross teaching between cnn and transformer. arXiv
**2021**, arXiv:2112.04894. [Google Scholar] - Liu, F.; Wei, H.; Zhao, W.; Li, G.; Peng, J.; Li, Z. WB-DETR: Transformer-based detector without backbone. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada, 10–17 October 2021; pp. 2979–2987. [Google Scholar]
- Chen, G.; Song, Z.; Qi, Z. Transformer-convolutional neural network for surface charge density profile prediction: Enabling high-throughput solvent screening with cosmo-sac. Chem. Eng. Sci.
**2021**, 246, 117002. [Google Scholar] [CrossRef] - Islam, M.S.; Hoque, M.E.; Amin, M.R. Integration of kalman filter in the epidemiological model: A robust approach to predict covid-19 outbreak in bangladesh. Int. J. Mod. Phys. C
**2021**, 32, 2150108. [Google Scholar] [CrossRef] - Yan, R.; Liu, J.; Wu, J.; Xu, C.; Hu, Y. The weak frequency anomaly detection method of atomic clocks based on kalman filter and extrapolation-accumulation. Meas. Control.
**2021**, 54, 565–575. [Google Scholar] [CrossRef] - Pannekoucke, O.; Ménard, R.; Aabaribaoune, M.E.; Plu, M. A methodology to obtain model-error covariances due to the discretization scheme from the parametric kalman filter perspective. Nonlinear Process. Geophys.
**2021**, 28, 1–22. [Google Scholar] [CrossRef] - Amjad, F.; Varanasi, S.K.; Huang, B. Kalman filter-based convolutional neural network for robust tracking of froth-middling interface in a primary separation vessel in presence of occlusions. IEEE Trans. Instrum. Meas.
**2021**, 70, 1–8. [Google Scholar] [CrossRef] - Ge, Z.; Liu, S.; Wang, F.; Li, Z.; Sun, J. YOLOX: Exceeding yolo series in 2021. arXiv
**2021**, arXiv:2107.08430. [Google Scholar] - Qin, Z.; Li, Z.; Zhang, Z.; Bao, Y.; Yu, G.; Peng, Y.; Sun, J. Thundernet: Towards real-time generic object detection on mobile devices. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Republic of Korea, 27 October–2 November 2019; pp. 6718–6727. [Google Scholar]
- Lu, K.; Xu, R.; Li, J.; Lv, Y.; Lin, H.; Liu, Y. A Vision-Based Detection and Spatial Localization Scheme for Forest Fire Inspection from UAV. Forests
**2022**, 13, 383. [Google Scholar] [CrossRef] - Chen, Z.; Yang, J.; Chen, L.; Jiao, H. Garbage classification system based on improved shufflenet v2. Resour. Conserv. Recycl.
**2022**, 178, 106090. [Google Scholar] [CrossRef] - Yin, X.; Li, W.; Li, Z.; Yi, L. Recognition of grape leaf diseases using MobileNetV3 and deep transfer learning. Int. J. Agric. Biol. Eng.
**2022**, 15, 184–194. [Google Scholar] [CrossRef]

**Figure 1.**Sample image frames: (

**a**) 1100 FPS, (

**b**) 800 FPS, (

**c**) 600 FPS, (

**d**) 400 FPS, (

**e**) 200 FPS, (

**f**) 120 FPS.

**Figure 7.**Example of a dataset: (

**a**) sample data taken horizontally to the desktop and (

**b**) sample data taken perpendicular to the desktop.

**Figure 8.**Network 1: (

**a**) configuration of Network 1 and (

**b**) test results for each parameter of Network 1.

**Figure 9.**Network 2: (

**a**) configuration of Network 2 and (

**b**) test results for each parameter of Network 2.

**Figure 10.**Network 3: (

**a**) configuration of Network 3 and (

**b**) test results for each parameter of Network 3.

**Figure 11.**Network 4: (

**a**) configuration of Network 4 and (

**b**) test results for each parameter of Network 4.

**Figure 12.**Network 5: (

**a**) configuration of Network 5 and (

**b**) test results for each parameter of Network 5.

**Figure 13.**Network 6: (

**a**) configuration of Network 6 and (

**b**) test results for each parameter of Network 6.

**Figure 15.**Comparison of target detection results using consecutive frames: (

**a**) GroundTruth, (

**b**) Network 1, (

**c**) Network 2, (

**d**) Network 3, (

**e**) Network 4, (

**f**) Network 5, (

**g**) Network 6.

**Figure 16.**Comparison results of different networks: (

**a**) 3D trajectory diagram of the sphere movement, (

**b**) trajectory of the sphere’s motion in x-coordinates, (

**c**) trajectory of the sphere in y-coordinate motion, (

**d**) trajectory diagram of the z-coordinates of the sphere’s motion.

Advantages | Disadvantages | Results | |
---|---|---|---|

Literature [7] | The ball flight model was established, and the BP pattern recognition classifier was employed, to identify patterns based on the predicted flight path. | The prediction ability of sphere trajectory was weak and the deviation was serious. | The accuracy of the trajectory prediction algorithm of the ball was above 92%. |

Literature [8] | A DCNN-LSTM (Deep Convolutional Neural Network Long Short-Term Memory) model was mainly used to improve the real-time performance of motion characteristics extraction through deep enhancement. In the network structure, DCNN was responsible for tracking and identifying objects, and the LSTM algorithm was responsible for predicting the trajectory of the ball. | The model did not have enough robustness and anti-interference ability and it was easy to lose the tracking target in the scene with complex background. | The mean and standard deviation of the error were 36.6 mm and 18.8 mm, respectively. |

Literature [9] | This article learned the parabolic trajectory of table tennis on the pivot of table tennis by building a neural network and realized the consequence of table tennis trajectory. | The model learned and predicted the overall trajectory, and the fit was good on the whole, but there were some deviations in the details. | The average error and standard deviation were 36.6 mm and 18.8 mm. |

Literature [10] | The automatic detection model was constructed by integrating a compensation fuzzy algorithm with a recursive neural network, resulting in a compensation fuzzy neural network algorithm. | The model was complex and needed to rely on high performance equipment for reasoning. | The precision of motion trajectory prediction improved, as the quantity of input data increased. A prediction error of less than 40 mm was achieved when utilizing 30 pieces of input data. |

Literature [11] | A table tennis target trajectory tracking algorithm combining machine vision and scale conjugate gradient was proposed to judge the rotating state of table tennis. | For the ball with fast hitting speed, it could not track the target effectively, and the target loss rate was higher. | The mean square error of the three-axis error of the neural network in this paper was 4.66. |

Epoch | Learning Rate (LR) | Momentum | Weight Decay | |
---|---|---|---|---|

1 | 300 | 0.0001 | 0.90 | 0.0005 |

2 | 300 | 0.0001 | 0.95 | 0.0005 |

3 | 500 | 0.0010 | 0.90 | 0.0005 |

4 | 500 | 0.0001 | 0.95 | 0.0005 |

5 | 700 | 0.0010 | 0.90 | 0.0005 |

6 | 700 | 0.0001 | 0.95 | 0.0005 |

Backbone Network | Return Module | Transformer Module | Lightweight Transformer Module | Kalman Filtering | Precision | Recall | AP | IoU | Parameter/M | FPS | |
---|---|---|---|---|---|---|---|---|---|---|---|

1 | √ | 85.8% | 81.6% | 77.2% | 70.6% | 1.23 | 935.67 | ||||

2 | √ | √ | 92.4% | 90.8% | 90.5% | 81.1% | 3.54 | 906.13 | |||

3 | √ | √ | √ | 96.0% | 94.6% | 91.7% | 83.3% | 14.77 | 348.73 | ||

4 | √ | √ | √ | 96.4% | 95.8% | 92.2% | 83.5% | 7.04 | 677.45 | ||

5 | √ | √ | √ | √ | 98.1% | 96.4% | 95.3% | 88.5% | 15.87 | 343.91 | |

6 | √ | √ | √ | √ | 98.2% | 97.3% | 96.8% | 89.1% | 7.68 | 634.19 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Li, W.; Liu, X.; An, K.; Qin, C.; Cheng, Y.
Table Tennis Track Detection Based on Temporal Feature Multiplexing Network. *Sensors* **2023**, *23*, 1726.
https://doi.org/10.3390/s23031726

**AMA Style**

Li W, Liu X, An K, Qin C, Cheng Y.
Table Tennis Track Detection Based on Temporal Feature Multiplexing Network. *Sensors*. 2023; 23(3):1726.
https://doi.org/10.3390/s23031726

**Chicago/Turabian Style**

Li, Wenjie, Xiangpeng Liu, Kang An, Chengjin Qin, and Yuhua Cheng.
2023. "Table Tennis Track Detection Based on Temporal Feature Multiplexing Network" *Sensors* 23, no. 3: 1726.
https://doi.org/10.3390/s23031726