Research on Automatic Evaluation Algorithm for Student Sports Movement Quality Based on Deep Learning
DOI:
https://doi.org/10.67541/jdc2602Keywords:
Movement quality assessment; Deep learning; Adaptive graph convolution; Temporal multi-scale aggregation; Contrastive learningAbstract
In sports teaching, the reliance on manual visual assessment of students' movement quality leads to subjectivity and low efficiency. This paper proposes an automatic assessment algorithm for sports movement quality based on deep learning. Firstly, a multimodal data platform is constructed to obtain RGB videos, depth skeletal sequences, and inertial measurement unit data. After time alignment and normalization, a three-dimensional skeletal motion sequence is generated. An adaptive spatio-temporal graph convolutional backbone network is designed. It dynamically models the collaborative relationships between non-adjacent joints via a learnable adjacency matrix and introduces a temporal multi-scale aggregation module to capture movement patterns at different time granularities. Furthermore, a hierarchical scoring regression module is proposed, which includes a global-local joint encoder, a segmented fine-grained scoring head, and a bias correction branch based on contrastive learning. It jointly models overall coordination, local posture, and individual difference compensation. Experiments on a self-built sports movement quality dataset show that the algorithm achieves a mean absolute error of 3.62 points, a Spearman rank correlation coefficient of 0.873, and a parameter count of 3.72 million. After pruning and TensorRT optimization, it performs real-time inference at 28.6 ms per sample on a Jetson Nano device. Ablation experiments and robustness tests verify the effectiveness of each core module, and the proposed method outperforms existing mainstream models in terms of accuracy and efficiency, providing a feasible technical solution for automatic movement assessment in sports teaching scenarios.
References
[1] Hu, Z., Liu, Z., & Su, Y. (2024). AI-driven smart transformation in physical education: Current trends and future research directions. Applied Sciences, 14(22), 10616. https://doi.org/10.3390/app142210616
[2] Sun, R., Liu, Y., Li, H., & Yim, J. (2026). A Study on the Factors Influencing User Experience of AI Pose Recognition Feedback Systems in Ballet-Class Contexts. Applied Sciences, 16(7), 3431. https://doi.org/10.3390/app16073431
[3] Van Maarseveen, M., Leenhouts, J., de Witte, A., Flux, E., Van Doorn, H., & van der Kamp, J. (2025). Enhancing affordance perception in pre-service physical education teachers: effects of content knowledge, motor experience and visual experience programs. Frontiers in Sports and Active Living, 7, 1583448. https://doi.org/10.3389/fspor.2025.1583448
[4] Feng, L., Zhao, Y., Zhao, W., & Tang, J. (2022). A comparative review of graph convolutional networks for human skeleton-based action recognition. Artificial Intelligence Review, 55(5), 4275-4305. https://doi.org/10.1007/s10462-021-10107-y
[5] Liu, Y., Zhang, H., Li, Y., He, K., & Xu, D. (2023). Skeleton-based human action recognition via large-kernel attention graph convolutional network. IEEE Transactions on Visualization and Computer Graphics, 29(5), 2575-2585. https://doi.org/10.1109/tvcg.2023.3247075
[6] Liu, F., Wang, C., Tian, Z., Du, S., & Zeng, W. (2025). Advancing skeleton-based human behavior recognition: multi-stream fusion spatiotemporal graph convolutional networks. Complex & Intelligent Systems, 11(1), 94. https://doi.org/10.1007/s40747-024-01743-2
[7] Zhang, C., Xu, Y., Xu, Z., Huang, J., & Lu, J. (2022). Hybrid handcrafted and learned feature framework for human action recognition: C. Zhang et al. Applied Intelligence, 52(11), 12771-12787. https://doi.org/10.1007/s10489-021-03068-w
[8] Sakaa, B., Elbeltagi, A., Boudibi, S., Chaffaï, H., Islam, A. R. M. T., Kulimushi, L. C., ... & Wong, Y. J. (2022). Water quality index modeling using random forest and improved SMO algorithm for support vector machine in Saf-Saf river basin. Environmental Science and Pollution Research, 29(32), 48491-48508. https://doi.org/10.1007/s11356-022-18644-x
[9] Adugna, T., Xu, W., & Fan, J. (2022). Comparison of random forest and support vector machine classifiers for regional land cover mapping using coarse resolution FY-3C images. Remote Sensing, 14(3), 574. https://doi.org/10.3390/rs14030574
[10] Pang, Y., Zhang, K., & Li, F. (2025). Explainable quality assessment of effective aligned skeletal representations for martial arts movements by multi-machine learning decisions. Scientific Reports, 15(1), 323. https://doi.org/10.1038/s41598-024-83475-4
[11] Freire-Obregón, D., Lorenzo-Navarro, J., Santana, O. J., Hernández-Sosa, D., & Castrillón-Santana, M. (2023, July). An x3d neural network analysis for runner’s performance assessment in a wild sporting environment. In 2023 18th International Conference on Machine Vision and Applications (MVA) (pp. 1-5). IEEE. https://doi.org/10.23919/mva57639.2023.10215918
[12] Bratta, C. (2024). Exploring Sex Differences in Diving Performance Analysis. https://hdl.handle.net/20.500.14242/190721
[13] Gan, Q. (2025). Sports Motion Analysis: From Competition Videos to Data-Driven Interpretations (Doctoral dissertation, Institut Polytechnique de Paris). https://doi.org/10.70675/5ae10807zc807z4f8dzb4d3z832fd95de4cb
[14] Yang, L., & Li, Y. (2026). Swimming action recognition algorithm based on improved C3D and attention-residual network. International Journal of Information and Communication Technology, 27(20), 40-62. https://doi.org/10.1504/ijict.2026.10076718
[15] Dong, K., Feng, X., & Dong, L. (2025, September). Dynamic Thermal Gesture Recognition Algorithm Based on C3D. In 2025 5th International Conference on Artificial Intelligence, Automation and High Performance Computing (AIAHPC) (pp. 126-133). IEEE. https://doi.org/10.1109/aiahpc66801.2025.11290620
[16] Lovanshi, M., & Tiwari, V. (2024). Human skeleton pose and spatio-temporal feature-based activity recognition using ST-GCN. Multimedia Tools and Applications, 83(5), 12705-12730.
https://doi.org/10.1007/s11042-023-16001-9
[17] Filtjens, B., Vanrumste, B., & Slaets, P. (2022). Skeleton-based action segmentation with multi-stage spatial-temporal graph convolutional neural networks. IEEE Transactions on Emerging Topics in Computing, 12(1), 202-212. https://doi.org/10.1109/tetc.2022.3230912
[18] Xu, J., Liu, F., Wang, Q., Zou, R., Wang, Y., Zheng, J., ... & Zeng, W. (2024). Enhancing human behavior recognition with spatiotemporal graph convolutional neural networks and skeleton sequences. EURASIP Journal on Advances in Signal Processing, 2024(1), 60. https://doi.org/10.1186/s13634-024-01156-w
[19] Zhou, C., Huang, Y., & Ling, H. (2022). Uncertainty-driven action quality assessment. arXiv preprint arXiv:2207.14513. https://doi.org/10.48550/arXiv.2207.14513
[20] Lei, Q., Yao, L., Zhang, H., & Du, J. (2025). Skeletal Spatio-Temporal Decoupling Transformer for Long-Duration Action Quality Assessment. Knowledge-Based Systems, 114672. https://doi.org/10.1016/j.knosys.2025.114672
[21] Zhu, S., Chen, J., & Su, Y. (2024). Spatio-temporal articulation & coordination co-attention graph network for human motion prediction. Signal Processing, 223, 109551. https://doi.org/10.1016/j.sigpro.2024.109551
[22] Shao, D., Zhao, Y., Dai, B., & Lin, D. (2020). Finegym: A hierarchical video dataset for fine-grained action understanding. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 2616-2625). https://doi.org/10.1109/cvpr42600.2020.00269
[23] Shuvo, M. M. H., Islam, S. K., Cheng, J., & Morshed, B. I. (2022). Efficient acceleration of deep learning inference on resource-constrained edge devices: A review. Proceedings of the IEEE, 111(1), 42-91. https://doi.org/10.1109/jproc.2022.3226481
[24] Sharma, D., & Sarkar, S. (2022). Enabling inference and training of deep learning models for AI applications on IoT edge devices. In Artificial Intelligence-based Internet of Things Systems (pp. 267-283). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-87059-1_10
[25] Shah, A., Bangash, J. I., Khan, A. W., Ahmed, I., Khan, A., Khan, A., & Khan, A. (2022). Comparative analysis of median filter and its variants for removal of impulse noise from gray scale images. Journal of King Saud University-Computer and Information Sciences, 34(3), 505-519. https://doi.org/10.1016/j.jksuci.2020.03.007
[26] Gao, Y., Dang, C., Zhu, J., Xie, Y., Hu, Y., Yan, C., ... & Li, X. (2025). A real-time 3D modeling method for buildings driven by IMU and RGB-D fusion. International Journal of Digital Earth, 18(1), 2506496. https://doi.org/10.1080/17538947.2025.2506496
[27] Lee, H., & Ryu, J. (2025). Toward Efficient Generalization in 3D Human Pose Estimation via a Canonical Domain Approach. IEEE Access. https://doi.org/10.1109/access.2025.3566109
[28] Zhang, J., Tu, Z., Weng, J., Yuan, J., & Du, B. (2024). A modular neural motion retargeting system decoupling skeleton and shape perception. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(10), 6889-6904. https://doi.org/10.1109/tpami.2024.3386777
[29] JunZhang, C., & GuanLi, Y. (2026). A hybrid framework combining adaptive graph learning and global temporal attention for skeleton-based action recognition. Scientific Reports. https://doi.org/10.1038/s41598-026-49915-z
[30] Iorga, A., Jianu, A., Gheorghiu, M., Crețu, B. D., & Eremia, I. A. (2023). Motor coordination and its importance in practicing performance movement. Sustainability, 15(7), 5812. https://doi.org/10.3390/su15075812
[31] Kilic, U., Karadag, O. O., & Ozyer, G. T. (2025). AGMS-GCN: Attention-guided multi-scale graph convolutional networks for skeleton-based action recognition. Knowledge-Based Systems, 311, 113045. https://doi.org/10.1016/j.knosys.2025.113045
[32] Singh, S. K., Bharti, B. K., Yadav, A. N., & Dwivedi, A. K. (2025). Optimized Microwave Ablation with a Novel Applicator: Integration of Taguchi Neural Networks for Enhanced Predictive Accuracy of Ablation Zone. IEEE Journal on Multiscale and Multiphysics Computational Techniques. https://doi.org/10.1109/jmmct.2025.3589163
[33] Tekin, M., Yurdal, M. O., Toraman, Ç., Korkmaz, G., & Uysal, İ. (2025). Is AI the future of evaluation in medical education?? AI vs. human evaluation in objective structured clinical examination. BMC medical education, 25(1), 641. https://doi.org/10.1186/s12909-025-07241-4
[34] Hu, K., Shen, C., Wang, T., Xu, K., Xia, Q., Xia, M., & Cai, C. (2024). Overview of temporal action detection based on deep learning. Artificial Intelligence Review, 57(2).
https://doi.org/10.1007/s10462-023-10650-w
[35] Zahan, S., Hassan, G. M., & Mian, A. (2024). Learning sparse temporal video mapping for action quality assessment in floor gymnastics. IEEE Transactions on Instrumentation and Measurement, 73, 1-11. https://doi.org/10.1109/tim.2024.3398072
[36] Wang, Y., Yue, Y., Lu, R., Han, Y., Song, S., & Huang, G. (2024). Efficienttrain++: Generalized curriculum learning for efficient visual backbone training. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12), 8036-8055. https://doi.org/10.48550/arXiv.2405.08768
Published
Data Availability Statement
支持本研究结果的数据可向通讯作者MZ索取。
Issue
Section
License
Copyright (c) 2026 Huicong Lei, Miao Zhou (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
