Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring
With the growing popularity of fitness, the demand for real-time action recognition and feedback is increasing. Current research faces challenges in handling complex actions, real-time processing, and system integration. To address these issues, we propose a novel fitness action recognition model th...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016824008019 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846113818184253440 |
|---|---|
| author | Jijie Li Ruyao Gong Gang Wang |
| author_facet | Jijie Li Ruyao Gong Gang Wang |
| author_sort | Jijie Li |
| collection | DOAJ |
| description | With the growing popularity of fitness, the demand for real-time action recognition and feedback is increasing. Current research faces challenges in handling complex actions, real-time processing, and system integration. To address these issues, we propose a novel fitness action recognition model that integrates ResNet, Transformer, and transfer learning techniques. Specifically, ResNet is used for image feature extraction, Transformer handles time-series data processing, and transfer learning accelerates the model’s adaptation to new data. We evaluated our model on the NTU RGB+D action recognition dataset, achieving 48.5 ms latency, 29.1 fps throughput, and 93.7% accuracy, significantly outperforming other models. Our model achieved an accuracy improvement of 5% over existing methods, demonstrating significant potential for real-time fitness monitoring. By incorporating IoT technology, our system enables real-time data processing and action recognition, making it ideal for smart fitness monitoring. Although the model has high complexity and memory usage, its efficiency and accuracy demonstrate its potential for widespread adoption. Future work will focus on optimizing the model structure and training methods to enhance applicability in resource-constrained environments, ensuring broader usability and efficiency in various real-world applications. |
| format | Article |
| id | doaj-art-d8567b628e004fd1ae4a258ba08d433a |
| institution | Kabale University |
| issn | 1110-0168 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Alexandria Engineering Journal |
| spelling | doaj-art-d8567b628e004fd1ae4a258ba08d433a2024-12-21T04:27:43ZengElsevierAlexandria Engineering Journal1110-01682024-12-0110989101Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoringJijie Li0Ruyao Gong1Gang Wang2Physical Education, Kunsan National University, Jeollabuk-do, Gunsan 54150, Republic of KoreaCollege of Sports and Health, Linyi University, Shandong, Linyi 276000, China; Corresponding author.School of Computing and Data Engineering, NingboTech University, Ningbo 315100, China; Department of Bioengineering, Imperial College London, London SW7 2AZ, UKWith the growing popularity of fitness, the demand for real-time action recognition and feedback is increasing. Current research faces challenges in handling complex actions, real-time processing, and system integration. To address these issues, we propose a novel fitness action recognition model that integrates ResNet, Transformer, and transfer learning techniques. Specifically, ResNet is used for image feature extraction, Transformer handles time-series data processing, and transfer learning accelerates the model’s adaptation to new data. We evaluated our model on the NTU RGB+D action recognition dataset, achieving 48.5 ms latency, 29.1 fps throughput, and 93.7% accuracy, significantly outperforming other models. Our model achieved an accuracy improvement of 5% over existing methods, demonstrating significant potential for real-time fitness monitoring. By incorporating IoT technology, our system enables real-time data processing and action recognition, making it ideal for smart fitness monitoring. Although the model has high complexity and memory usage, its efficiency and accuracy demonstrate its potential for widespread adoption. Future work will focus on optimizing the model structure and training methods to enhance applicability in resource-constrained environments, ensuring broader usability and efficiency in various real-world applications.http://www.sciencedirect.com/science/article/pii/S1110016824008019Action recognitionImage analysisResNetTransformerTransfer learning |
| spellingShingle | Jijie Li Ruyao Gong Gang Wang Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring Alexandria Engineering Journal Action recognition Image analysis ResNet Transformer Transfer learning |
| title | Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring |
| title_full | Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring |
| title_fullStr | Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring |
| title_full_unstemmed | Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring |
| title_short | Enhancing fitness action recognition with ResNet-TransFit: Integrating IoT and deep learning techniques for real-time monitoring |
| title_sort | enhancing fitness action recognition with resnet transfit integrating iot and deep learning techniques for real time monitoring |
| topic | Action recognition Image analysis ResNet Transformer Transfer learning |
| url | http://www.sciencedirect.com/science/article/pii/S1110016824008019 |
| work_keys_str_mv | AT jijieli enhancingfitnessactionrecognitionwithresnettransfitintegratingiotanddeeplearningtechniquesforrealtimemonitoring AT ruyaogong enhancingfitnessactionrecognitionwithresnettransfitintegratingiotanddeeplearningtechniquesforrealtimemonitoring AT gangwang enhancingfitnessactionrecognitionwithresnettransfitintegratingiotanddeeplearningtechniquesforrealtimemonitoring |