Table Tennis Track Detection Based on Temporal Feature Multiplexing Network
Recording the trajectory of table tennis balls in real-time enables the analysis of the opponent’s attacking characteristics and weaknesses. The current analysis of the ball paths mainly relied on human viewing, which lacked certain theoretical data support. In order to solve the problem of the lack...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/3/1726 |
| _version_ | 1797623131792211968 |
|---|---|
| author | Wenjie Li Xiangpeng Liu Kang An Chengjin Qin Yuhua Cheng |
| author_facet | Wenjie Li Xiangpeng Liu Kang An Chengjin Qin Yuhua Cheng |
| author_sort | Wenjie Li |
| collection | DOAJ |
| description | Recording the trajectory of table tennis balls in real-time enables the analysis of the opponent’s attacking characteristics and weaknesses. The current analysis of the ball paths mainly relied on human viewing, which lacked certain theoretical data support. In order to solve the problem of the lack of objective data analysis in the research of table tennis competition, a target detection algorithm-based table tennis trajectory extraction network was proposed to record the trajectory of the table tennis movement in video. The network improved the feature reuse rate in order to achieve a lightweight network and enhance the detection accuracy. The core of the network was the “feature store & return” module, which could store the output of the current network layer and pass the features to the input of the network layer at the next moment to achieve efficient reuse of the features. In this module, the Transformer model was used to secondarily process the features, build the global association information, and enhance the feature richness of the feature map. According to the designed experiments, the detection accuracy of the network was 96.8% for table tennis and 89.1% for target localization. Moreover, the parameter size of the model was only 7.68 MB, and the detection frame rate could reach 634.19 FPS using the hardware for the tests. In summary, the network designed in this paper has the characteristics of both lightweight and high precision in table tennis detection, and the performance of the proposed model significantly outperforms that of the existing models. |
| first_indexed | 2024-03-11T09:24:20Z |
| format | Article |
| id | doaj.art-9e09edc2a4a2462c94ddebc0186a9cfb |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T09:24:20Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-9e09edc2a4a2462c94ddebc0186a9cfb2023-11-16T18:05:14ZengMDPI AGSensors1424-82202023-02-01233172610.3390/s23031726Table Tennis Track Detection Based on Temporal Feature Multiplexing NetworkWenjie Li0Xiangpeng Liu1Kang An2Chengjin Qin3Yuhua Cheng4College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaCollege of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 201418, ChinaSchool of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, ChinaShanghai Research Institute of Microelectronics, Peking University, Shanghai 201203, ChinaRecording the trajectory of table tennis balls in real-time enables the analysis of the opponent’s attacking characteristics and weaknesses. The current analysis of the ball paths mainly relied on human viewing, which lacked certain theoretical data support. In order to solve the problem of the lack of objective data analysis in the research of table tennis competition, a target detection algorithm-based table tennis trajectory extraction network was proposed to record the trajectory of the table tennis movement in video. The network improved the feature reuse rate in order to achieve a lightweight network and enhance the detection accuracy. The core of the network was the “feature store & return” module, which could store the output of the current network layer and pass the features to the input of the network layer at the next moment to achieve efficient reuse of the features. In this module, the Transformer model was used to secondarily process the features, build the global association information, and enhance the feature richness of the feature map. According to the designed experiments, the detection accuracy of the network was 96.8% for table tennis and 89.1% for target localization. Moreover, the parameter size of the model was only 7.68 MB, and the detection frame rate could reach 634.19 FPS using the hardware for the tests. In summary, the network designed in this paper has the characteristics of both lightweight and high precision in table tennis detection, and the performance of the proposed model significantly outperforms that of the existing models.https://www.mdpi.com/1424-8220/23/3/1726motion trajectorydeep learningobject detectionlightweight networkTransformer modelfeature reuse |
| spellingShingle | Wenjie Li Xiangpeng Liu Kang An Chengjin Qin Yuhua Cheng Table Tennis Track Detection Based on Temporal Feature Multiplexing Network Sensors motion trajectory deep learning object detection lightweight network Transformer model feature reuse |
| title | Table Tennis Track Detection Based on Temporal Feature Multiplexing Network |
| title_full | Table Tennis Track Detection Based on Temporal Feature Multiplexing Network |
| title_fullStr | Table Tennis Track Detection Based on Temporal Feature Multiplexing Network |
| title_full_unstemmed | Table Tennis Track Detection Based on Temporal Feature Multiplexing Network |
| title_short | Table Tennis Track Detection Based on Temporal Feature Multiplexing Network |
| title_sort | table tennis track detection based on temporal feature multiplexing network |
| topic | motion trajectory deep learning object detection lightweight network Transformer model feature reuse |
| url | https://www.mdpi.com/1424-8220/23/3/1726 |
| work_keys_str_mv | AT wenjieli tabletennistrackdetectionbasedontemporalfeaturemultiplexingnetwork AT xiangpengliu tabletennistrackdetectionbasedontemporalfeaturemultiplexingnetwork AT kangan tabletennistrackdetectionbasedontemporalfeaturemultiplexingnetwork AT chengjinqin tabletennistrackdetectionbasedontemporalfeaturemultiplexingnetwork AT yuhuacheng tabletennistrackdetectionbasedontemporalfeaturemultiplexingnetwork |