DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning
With the rapid development of IoT technology, real-time human pose estimation has become increasingly important in sports training feedback systems. However, current methods often fall short in balancing high accuracy with low computational resource requirements, especially in resource-constrained e...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016824011657 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583075530276864 |
|---|---|
| author | Rongbao Huang Bo Zhang Zhixin Yao Bojun Xie Jia Guo |
| author_facet | Rongbao Huang Bo Zhang Zhixin Yao Bojun Xie Jia Guo |
| author_sort | Rongbao Huang |
| collection | DOAJ |
| description | With the rapid development of IoT technology, real-time human pose estimation has become increasingly important in sports training feedback systems. However, current methods often fall short in balancing high accuracy with low computational resource requirements, especially in resource-constrained environments. Deep learning has shown significant potential in enhancing computer vision tasks, including human pose estimation. In this study, we propose DESNet, an improved EfficientHRNet model that integrates IoT technology. DESNet combines Dynamic Multi-Scale Context (DMC) modules and Squeeze-and-Excitation (SE) modules, and utilizes IoT for real-time data collection, transmission, and processing. Experimental results show that DESNet achieves an average precision (AP) of 74.8% on the COCO dataset and a PCKh (Percentage of Correct Keypoints with head-normalized) of 90.9% on the MPII dataset, outperforming existing lightweight models. The integration of deep learning and IoT technology not only improves the accuracy and efficiency of human pose estimation but also significantly enhances the timeliness and robustness of feedback in sports training applications. Our findings demonstrate that DESNet is a powerful tool for real-time human pose analysis, offering promising solutions for intelligent sports training and rehabilitation systems. |
| format | Article |
| id | doaj-art-2cb8c26263bc492e9080c6e336800de4 |
| institution | Kabale University |
| issn | 1110-0168 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Alexandria Engineering Journal |
| spelling | doaj-art-2cb8c26263bc492e9080c6e336800de42025-01-29T05:00:03ZengElsevierAlexandria Engineering Journal1110-01682025-01-01112293306DESNet: Real-time human pose estimation for sports applications combining IoT and deep learningRongbao Huang0Bo Zhang1Zhixin Yao2Bojun Xie3Jia Guo4Department of Physical Education and Teaching, Hebei Finance University, Baoding, 071000, ChinaDepartment of Physical Education and Teaching, Hebei Finance University, Baoding, 071000, ChinaSchool of Physical Education and Health, LinYi University, Lanshan district, Linyi City, Shandong province, 276000, China; Department of Sports and Leisure Services, Pai Chai University, Daejeon, 35345, South Korea; Corresponding author at: Department of Sports and Leisure Services, Pai Chai University, Daejeon, 35345, South Korea.College of Mathematics and Information Science of Hebei University, Baoding, Hebei, 071000, ChinaSchool of Information Engineering and Computer Science, Hebei Finance University, Baoding, Hebei, 071000, ChinaWith the rapid development of IoT technology, real-time human pose estimation has become increasingly important in sports training feedback systems. However, current methods often fall short in balancing high accuracy with low computational resource requirements, especially in resource-constrained environments. Deep learning has shown significant potential in enhancing computer vision tasks, including human pose estimation. In this study, we propose DESNet, an improved EfficientHRNet model that integrates IoT technology. DESNet combines Dynamic Multi-Scale Context (DMC) modules and Squeeze-and-Excitation (SE) modules, and utilizes IoT for real-time data collection, transmission, and processing. Experimental results show that DESNet achieves an average precision (AP) of 74.8% on the COCO dataset and a PCKh (Percentage of Correct Keypoints with head-normalized) of 90.9% on the MPII dataset, outperforming existing lightweight models. The integration of deep learning and IoT technology not only improves the accuracy and efficiency of human pose estimation but also significantly enhances the timeliness and robustness of feedback in sports training applications. Our findings demonstrate that DESNet is a powerful tool for real-time human pose analysis, offering promising solutions for intelligent sports training and rehabilitation systems.http://www.sciencedirect.com/science/article/pii/S1110016824011657Real-time human pose estimationSports training feedbackIoT integrationDESNetDynamic Multi-Scale ContextSqueeze-and-Excitation |
| spellingShingle | Rongbao Huang Bo Zhang Zhixin Yao Bojun Xie Jia Guo DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning Alexandria Engineering Journal Real-time human pose estimation Sports training feedback IoT integration DESNet Dynamic Multi-Scale Context Squeeze-and-Excitation |
| title | DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning |
| title_full | DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning |
| title_fullStr | DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning |
| title_full_unstemmed | DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning |
| title_short | DESNet: Real-time human pose estimation for sports applications combining IoT and deep learning |
| title_sort | desnet real time human pose estimation for sports applications combining iot and deep learning |
| topic | Real-time human pose estimation Sports training feedback IoT integration DESNet Dynamic Multi-Scale Context Squeeze-and-Excitation |
| url | http://www.sciencedirect.com/science/article/pii/S1110016824011657 |
| work_keys_str_mv | AT rongbaohuang desnetrealtimehumanposeestimationforsportsapplicationscombiningiotanddeeplearning AT bozhang desnetrealtimehumanposeestimationforsportsapplicationscombiningiotanddeeplearning AT zhixinyao desnetrealtimehumanposeestimationforsportsapplicationscombiningiotanddeeplearning AT bojunxie desnetrealtimehumanposeestimationforsportsapplicationscombiningiotanddeeplearning AT jiaguo desnetrealtimehumanposeestimationforsportsapplicationscombiningiotanddeeplearning |