The week's pick
Random Articles
Reseach Article
AI-Enabled Reform of University Physical Education Curriculum under Modern Pedagogical Paradigms: A Deep-Learning–Driven Intelligent Teaching Framework
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 187 - Number 77 |
| Year of Publication: 2026 |
| Authors: Bin Yan |
10.5120/ijca2026926255
|
Bin Yan . AI-Enabled Reform of University Physical Education Curriculum under Modern Pedagogical Paradigms: A Deep-Learning–Driven Intelligent Teaching Framework. International Journal of Computer Applications. 187, 77 ( Jan 2026), 9-15. DOI=10.5120/ijca2026926255
Abstract
With the rapid advancement of artificial intelligence (AI) and deep learning, higher education is undergoing a paradigm shift toward intelligent, individualized, and data-driven instructional models. As a core component of university curricula, physical education (PE) must similarly evolve to support precise skill acquisition, personalized training, and objective performance assessment. However, traditional PE instruction often relies on subjective observation, uniform training structures, and limited formative feedback, constraining student engagement, motor-skill development, and learning efficiency. To address these limitations, this study investigates an AI-driven reform pathway for university PE and proposes an integrated intelligent PE framework that combines deep-learning–based human motion analysis, automated feedback mechanisms, and data-driven personalized training plans. The system leverages pose-estimation models and multi-dimensional motion features to evaluate movement quality, track physical literacy development, and generate individualized corrective guidance in real time. Edge-enhanced inference and privacy-preserving data pipelines ensure deployability in real campus environments. Experimental evaluation across benchmark datasets and university pilot scenarios demonstrates that the proposed framework substantially improves motor-skill recognition accuracy, movement-quality scoring, and learning-progress stability, achieving up to +8.5% accuracy, +15.9% biomechanical quality, and +18.3% progression improvement over competitive baselines.
References
- AHMED, S., AND KUMAR, R. Teacher readiness and technological literacy for ai integration in university pe programs. Teaching and Teacher Education (2024).
- CAO, Z., SIMON, T., WEI, S.-E., AND SHEIKH, Y. Realtime multi-person 2d pose estimation using part affinity fields. In CVPR (2017).
- CHEN, L., ET AL. Artificial intelligence–assisted physical education: Adaptive learning and performance feedback. Computers & Education (2022).
- CHEN, Y., ET AL. Channel-wise topology refining graph convolution network. In ICCV (2021).
- DUAN, H., ET AL. Revisiting skeleton-based action recognition. In CVPR (2022).
- ET AL., A. C. S. Wearable technology in sports: Challenges and opportunities. Appl. Sci. (2023).
- ET AL., A. D. An image is worth 16x16 words. arXiv preprint (2020).
- ET AL., A. M. Deeplabcut: markerless pose estimation of user-defined body parts. Nat. Neurosci. (2018).
- ET AL., B. Y. Federated learning: applications and tools. Heliyon (2024).
- ET AL., D. Z. Artificial intelligence in sport: A narrative review. J. Sports Sci. (2025).
- ET AL., H. G. Physical education analytics with mcdm. Sci. Rep. (2025).
- ET AL., J. B. Ai in physical education: A systematic review. Apunts Educ. Fis. Deportes (2024).
- ET AL., K. S. Deep high-resolution representation learning for human pose estimation. In CVPR (2019).
- ET AL., P. K. Advances and open problems in federated learning. Found. Trends Mach. Learn. (2021).
- ET AL., T. D. Privacy-preserving in-bed pose monitoring. arXiv preprint (2022).
- ET AL., T.-Y. L. Privacy-enhanced representation for pose estimation. In BMVC (2023).
- ET AL., V. B. Blazepose: On-device real-time body pose tracking. arXiv preprint (2020).
- ET AL., X. J. Survey on edge computing for wearable technology. Computer Communications (2022).
- ET AL., Y. D. Edge digital sports management. J. Cloud Computing (2023).
- ET AL., Y. W. Ai in physical education: Teacher development perspective. Front. Psychol. (2024).
- FANG, H.-S., XIE, S., TAI, Y.-W., AND LU, C. Rmpe: Regional multi-person pose estimation. In ICCV (2016).
- IONESCU, C., PAPAVA, D., OLARU, V., AND SMINCHISESCU, C. Human3.6m: Large scale datasets and predictive methods for 3d human sensing.
- KAY, W. E. A. The kinetics human action video dataset. In CVPR (2017).
- KIM, D., AND PARK, J. Self-supervised human motion evaluation for biomechanics and athletic training. IEEE Transactions on Multimedia (2023).
- KUEHNE, H., JHUANG, H., GARROTE, E., POGGIO, T., AND SERRE, T. Hmdb: A large video database for human motion recognition. In ICCV (2011).
- LIU, H., CHEN, Y., AND ZHAO, Q. Biomechanics-informed deep motion analysis for fair and robust skill evaluation in physical education. IEEE Transactions on Affective Computing (2024).
- LIU, J., ET AL. Ntu rgb+d 120: A large-scale benchmark for 3d human activity understanding. TPAMI (2019).
- LOPEZ, M., ZHANG, Q., AND HASSAN, T. Deep learning– enabled skill assessment in university sports training. IEEE Transactions on Learning Technologies (2023).
- MMLAB. Rtmpose: Real-time multi-person pose estimation. arXiv preprint (2023).
- PELLICER, M., AND CHEN, Y. Edge intelligence and hybrid learning environments for ai-enabled physical education. Future Generation Computer Systems (2025).
- ROSSI, A., AND FERRI, G. Learning analytics models for physical literacy and performance evaluation in higher education. British Journal of Educational Technology (2023).
- SHAHROUDY, A., LIU, T., NG, T.-Y., AND WANG, G. Ntu rgb+d: A large scale dataset for 3d human activity analysis. In CVPR (2016).
- SINGH, A., WANG, L., AND ZHAO, M. Federated learning frameworks for privacy-preserving student motion data in sports education. IEEE Internet of Things Journal (2024).
- TANG, Z., ZHAO, T., ZHANG, T., PHAN, H., WANG, Y., SHI, C., YUAN, B., AND CHEN, Y. Rf domain backdoor attack on signal classification via stealthy trigger. IEEE Transactions on Mobile Computing 23, 12 (2024), 11765–11780.
- WANG, C.-Y., YEH, I.-H., AND LIAO, H.-Y. Yolov7: Trainable bag-of-freebies. In arXiv preprint (2022).
- WU, H., SINGH, P., AND LI, F. Hybrid kinematic-temporal deep networks for high-precision sports pose analysis. In CVPR Workshops (2024).
- YAN, S., XIONG, Y., AND LIN, D. Spatial temporal graph convolutional networks for skeleton-based action recognition. In AAAI (2018).
- ZHANG, Y., AND LIU, J. Transpose-sport: Transformerbased pose estimation for athletic performance understanding. Pattern Recognition Letters (2022).
Index Terms
Keywords