Action Quality Assessment Based on Kinematic Reconstruction and Deviation Analysis
Source abstract
Abstract
Action quality assessment (AQA) is vital in sports science and clinical rehabilitation, but existing methods suffer from data noise sensitivity, weak physical interpretability, and limited generalization across body types. This study proposes a novel AQA paradigm for standing long jump to address these limitations. A methodological framework based on kinematic reconstruction and physical deviation analysis is developed. Ideal trajectories derived from forward kinematics establish a unified evaluation benchmark. Three deviation features (Peak Deviation, Cumulative Deviation Integral, Phase-Weighted Deviation) are computed between ideal and experimental samples, then input into a Random Forest Classifier for automatic grading. Experimental results demonstrate the proposed method significantly outperforms traditional dynamic-time-warping-based templates in accuracy and robustness. Ablation studies confirm kinematic trajectory benchmarks are superior to raw or averaged trajectories. Model decisions show strong consistency with physical characteristics, verifying interpretable assessment capability. This work shifts AQA from “data similarity comparison” to “physical rationality evaluation”, offering a new perspective for vision-based automated motion analysis.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Balance, Gait, and Falls Prevention · Human Pose and Action Recognition · Sports Performance and Training
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Engineering Science and Technology, Shanghai Ocean University, Shanghai 201306, China
School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, China
References (41)
Show source-provided references
1. Zhou, K., Cai, R., Wang, L., Hubert, S., and Liang, X. (2024). A Comprehensive Survey of Action Quality Assessment: Method and Benchmark. arXiv.
2. Pose-guided matching based on deep learning for assessing quality of action on rehabilitation training10.1016/j.bspc.2021.103323 ↗
3. Lei, Q., Du, J., Zhang, H., Ye, S., and Chen, D. (2019). A Survey of Vision-Based Human Action Evaluation Methods. Sensors, 19.10.3390/s19194129 ↗
4. Multi-skeleton structures graph convolutional network for action quality assessment in long videos10.1007/s10489-023-04613-5 ↗
5. Wang, X., Li, J., and Hu, H. (2022). Skeleton-Based Action Quality Assessment via Partially Connected LSTM with Triplet Losses. Chinese Conference on Pattern Recognition and Computer Vision (PRCV), Springer Nature.10.1007/978-3-031-18913-5_17 ↗
6. Cao, Z., Simon, T., Wei, S., and Sheikh, Y. (2017). Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017, IEEE.10.1109/CVPR.2017.143 ↗
7. Fang, H., Xie, S., Tai, Y., and Lu, C. (2017, January 22–29). RMPE: Regional Multi-Person Pose Estimation. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.10.1109/ICCV.2017.256 ↗
8. Wei, S., Ramakrishna, V., Kanade, T., and Sheikh, Y. (2016). Convolutional pose machines. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016, IEEE.10.1109/CVPR.2016.511 ↗
9. Sun, K., Xiao, B., Liu, D., and Wang, J. (2019). Deep high-resolution representation learning for human pose estimation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, 16–20 June 2019, IEEE.10.1109/CVPR.2019.00584 ↗
10. Liu, J., Hou, Q., Cheng, M., Wang, C., and Feng, J. (2020). Improving convolutional networks with self-calibrated convolutions. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA, 13–19 June 2020, IEEE.10.1109/CVPR42600.2020.01011 ↗
11. Skeleton-based deep pose feature learning for action quality assessment on figure skating videos10.1016/j.jvcir.2022.103625 ↗
12. Wang, S., Yang, D., Zhai, P., Yu, Q., Suo, T., Sun, Z., Li, K., and Zhang, L. (2021). A survey of video-based action quality assessment. Proceedings of the 2021 International Conference on Networking Systems of AI, Shanghai, China, 19–20 November 2021, IEEE.10.1109/INSAI54028.2021.00029 ↗
13. Attention-driven multimodal alignment for long-term action quality assessment10.1016/j.asoc.2025.113649 ↗
14. Human-centric fine-grained action quality assessment10.1109/TPAMI.2025.3556935 ↗
15. Pose-guided transformer for fine-grained action quality assessment10.1109/TCSVT.2025.3545625 ↗
16. I3D-AE-LSTM: Combining action representations using a 2-stream autoencoder for action quality assessment10.1016/j.eswa.2025.127368 ↗
17. Multimodal action quality assessment10.1109/TIP.2024.3362135 ↗
18. Vision-based human action quality assessment: A systematic review10.1016/j.eswa.2024.125642 ↗
19. ResFNN: Residual Structure-Based Feedforward Neural Network for Action Quality Assessment in Sports Consumer Electronics10.1109/TCE.2024.3482560 ↗
20. Hierarchical Graph Convolutional Networks for Action Quality Assessment10.1109/TCSVT.2023.3281413 ↗
21. Davis, J., and Bobick, A. (1997). The Representation and Recognition of Human Movement Using Temporal Templates. Proceedings of the Computer Vision and Pattern Recognition, San Juan, PR, USA, 17–19 June 1997, IEEE Computer Society.10.1109/CVPR.1997.609439 ↗
22. Motion history image: Its variants and applications10.1007/s00138-010-0298-4 ↗
23. Parmar, P., and Morris, B.T. (2019, January 16–20). What and How Well You Performed? A Multitask Learning Approach to Action Quality Assessment. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00039 ↗
24. Parmar, P., and Morris, B.T. (2019, January 7–11). Action Quality Assessment Across Multiple Actions. Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV), Waikoloa Village, HI, USA.10.1109/WACV.2019.00161 ↗
25. Parmar, P., and Morris, B.T. (2017, January 21–26). Learning To Score Olympic Events. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, HI, USA.10.1109/CVPRW.2017.16 ↗
26. Personal Rehabilitation Exercise Assistant with Kinect and Dynamic Time Warping
27. Interpretable two-stage action quality assessment via 3D human pose estimation and dynamic feature alignment10.1007/s00371-025-04101-6 ↗
28. Lei, Q., Zhang, H., Du, J., Hsiao, T., and Chen, C. (2020). Learning effective skeletal representations on RGB video for fine-grained human action quality assessment. Electronics, 9.10.3390/electronics9040568 ↗
29. A Comparative Study of Pose Representation and Dynamics Modelling for Online Motion Quality Assessment10.1016/j.cviu.2015.11.016 ↗
30. Som, A., Anirudh, R., Wang, Q., and Turaga, P. (2016). Riemannian Geometric Approaches for Measuring Movement Quality. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Las Vegas, NV, USA, 26 June–1 July 2016, IEEE Computer Society.10.1109/CVPRW.2016.129 ↗
31. Stone, E., Skubic, M., and Back, J. (2014). Automated Health Alerts from Kinect-Based In-Home Gait Measurements. Proceedings of the 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, 26–30 August 2014, IEEE.10.1109/EMBC.2014.6944244 ↗
32. Rocha, A., Choupina, H., Fernandes, J., Rosas, M., Vaz, R., and Cunha, J. (2015). Kinect V2 Based System for Parkinson’s Disease Assessment. Proceedings of the 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Milan, Italy, 25–29 August 2015, IEEE.10.1109/EMBC.2015.7318601 ↗
33. Motion Tracking and Gait Feature Estimation for Recognising Parkinson’s Disease Using MS Kinect10.1186/s12938-015-0092-7 ↗
34. Bogo, F., Kanazawa, A., Lassner, C., Gehler, P., Romero, J., and Black, M.J. (2016, January 11–14). Keep It SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image. Proceedings of the European Conference on Computer Vision (ECCV), Amsterdam, The Netherlands.10.1007/978-3-319-46454-1_34 ↗
35. Kanazawa, A., Black, M.J., Jacobs, D.W., and Malik, J. (2018, January 18–22). End-to-End Recovery of Human Shape and Pose. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA.10.1109/CVPR.2018.00744 ↗
36. Kocabas, M., Athanasiou, N., and Black, M.J. (2020, January 13–19). VIBE: Video Inference for Human Body Pose and Shape Estimation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.10.1109/CVPR42600.2020.00530 ↗
37. Choi, H., Moon, G., and Lee, K.M. (2020, January 23–28). Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose. Proceedings of the European Conference on Computer Vision (ECCV), Glasgow, UK.10.1007/978-3-030-58571-6_45 ↗
38. Pavllo, D., Feichtenhofer, C., Grangier, D., and Auli, M. (2019, January 16–20). 3D Human Pose Estimation in Video with Temporal Convolutions and Semi-Supervised Training. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00794 ↗
39. Xie, K., Wang, T., Iqbal, U., Guo, Y., Fidler, S., and Shkurti, F. (2021, January 11–17). Physics-Based Human Motion Estimation and Synthesis from Videos. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada.10.1109/ICCV48922.2021.01133 ↗
40. Normal Range of Motion of Joints in Male Subjects10.2106/00004623-197961050-00017 ↗
41. Macedo, L.G. (2007). Normal Range of Motion of Joints in Female Subjects in Different Age Groups. [Master’s Thesis, University of Alberta].