{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,9]],"date-time":"2026-01-09T07:31:29Z","timestamp":1767943889689,"version":"3.49.0"},"reference-count":28,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2026,1,7]],"date-time":"2026-01-07T00:00:00Z","timestamp":1767744000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002628","name":"Incheon National University","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100002628","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Accurate golf swing correction requires modeling not only static pose deviations but also temporal rhythm and biomechanical stability throughout the swing sequence. Most existing pose-based approaches rely on frame-wise similarity and therefore fail to capture timing, velocity transitions, and coordinated joint dynamics. This study proposes a reinforcement learning-based framework that generates frame-level corrective motions by formulating swing correction as a sequential decision-making problem optimized via Proximal Policy Optimization (PPO). A multi-term reward function is designed to integrate geometric pose accuracy, incremental correction improvement, hip-centered stability, and temporal rhythm consistency measured using a Velocity-DTW metric. Experiments conducted with swing sequences from one professional and five amateur golfers demonstrate that the proposed method produces smoother and more temporally coherent corrections than static pose\u2013based baselines. In particular, rhythm-aware rewards substantially improve the motion of highly dynamic joints, such as the wrists and shoulders, while preserving lower-body stability. Visual analyses further confirm that the corrected trajectories follow expert patterns in both spatial alignment and timing. These results indicate that explicitly incorporating temporal rhythm within a reinforcement learning framework is essential for realistic and effective swing correction. The proposed method provides a principled foundation for automated, expert-level coaching systems in golf and other dynamic sports requiring temporally coordinated whole-body motion.<\/jats:p>","DOI":"10.3390\/s26020392","type":"journal-article","created":{"date-parts":[[2026,1,7]],"date-time":"2026-01-07T14:02:51Z","timestamp":1767794571000},"page":"392","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Reinforcement Learning-Based Golf Swing Correction Framework Incorporating Temporal Rhythm and Kinematic Stability"],"prefix":"10.3390","volume":"26","author":[{"given":"Dong-Jun","family":"Lee","sequence":"first","affiliation":[{"name":"Department of Embedded Systems Engineering, College of Information Technology, Incheon National University, Incheon 22012, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Young-Been","family":"Noh","sequence":"additional","affiliation":[{"name":"Department of Embedded Systems Engineering, College of Information Technology, Incheon National University, Incheon 22012, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2563-4443","authenticated-orcid":false,"given":"Jeongeun","family":"Byun","sequence":"additional","affiliation":[{"name":"Research Center for Technology Commercialization, Korea Institute of Science & Technology Information (KISTI), Seoul 02456, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4196-4725","authenticated-orcid":false,"given":"Kwang-Il","family":"Hwang","sequence":"additional","affiliation":[{"name":"Department of Embedded Systems Engineering, College of Information Technology, Incheon National University, Incheon 22012, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,1,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Crabbe, B., Perreault, S., Roy, S., and Ferrie, F.P. (2015, January 7\u201313). Skeleton-Free Body Pose Estimation from Depth Images for Movement Analysis. Proceedings of the IEEE International Conference on Computer Vision Workshops (ICCVW), Santiago, Chile.","DOI":"10.1109\/ICCVW.2015.49"},{"key":"ref_2","unstructured":"Mori, G., Tran, M., Wu, Y.N., and Zhu, S.C. (2015). Pose Embeddings: A Deep Architecture for Learning to Match Human Poses. arXiv."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"59","DOI":"10.55708\/js0103007","article-title":"An Evaluation of 2D Human Pose Estimation Based on ResNet Backbone","volume":"1","author":"Bui","year":"2022","journal-title":"J. Eng. Res. Sci."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"172","DOI":"10.1109\/TPAMI.2019.2929257","article-title":"OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields","volume":"43","author":"Cao","year":"2019","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_5","unstructured":"Bazarevsky, V., Grishchenko, I., Raveendran, K., Zhang, F., and Grundmann, M. (2020). BlazePose: On-Device Real-Time Body Pose Tracking. arXiv."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Geng, Z., Sun, K., Xiao, B., Zhao, T., Yang, J., Wang, C., and Cheng, X. (2021, January 19\u201325). Bottom-Up Human Pose Estimation via Disentangled Keypoint Regression. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.","DOI":"10.1109\/CVPR46437.2021.01444"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Liu, J., Wang, H., Tan, J., and He, M. (2021, January 10\u201317). Normalized Human Pose Features for Human Action Video Alignment. Proceedings of the IEEE\/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada.","DOI":"10.1109\/ICCV48922.2021.01132"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"30908","DOI":"10.1038\/s41598-024-81762-8","article-title":"Human Motion Similarity Evaluation Based on Deep Metric Learning","volume":"14","author":"Zhang","year":"2024","journal-title":"Sci. Rep."},{"key":"ref_9","unstructured":"Collins, R.T. (2023). A Lightweight Contrastive Approach for Aligning Human Pose Sequences. arXiv."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Ju, C.-Y., Kim, J.-H., and Lee, D.-H. (2023). GolfMate: Enhanced Golf Swing Analysis Tool through Pose Refinement Network and Explainable Golf Swing Embedding for Self-Training. Appl. Sci., 13.","DOI":"10.3390\/app132011227"},{"key":"ref_11","first-page":"39","article-title":"The Methodology of the Golf Swing Similarity Measurement Using Deep Learning-Based 2D Pose Estimation","volume":"28","author":"Park","year":"2023","journal-title":"J. Korea Soc. Comput. Inf."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Lee, M.-H., Yu, Y., Cho, S., Kim, J.-H., and Park, S. (2024, January 3\u20138). GolfPose: From Regular Posture to Golf Swing Posture. Proceedings of the International Conference on Pattern Recognition (ICPR), Kolkata, India.","DOI":"10.1007\/978-3-031-78305-0_25"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"126","DOI":"10.1186\/s12984-021-00919-y","article-title":"Deep Reinforcement Learning for Modeling Human Locomotion Control in Neuromechanical Simulation","volume":"18","author":"Song","year":"2021","journal-title":"J. Neuroeng. Rehabil."},{"key":"ref_14","first-page":"4023861","article-title":"Human Motion Posture Detection Algorithm Using Deep Reinforcement Learning","volume":"2021","author":"Qi","year":"2021","journal-title":"Mobile Inf. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Peters, J., and Schaal, S. (2006, January 9\u201315). Policy Gradient Methods for Robotics. Proceedings of the IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS), Beijing, China.","DOI":"10.1109\/IROS.2006.282564"},{"key":"ref_16","unstructured":"Schulman, J., Wolski, F., Dhariwal, P., Radford, A., and Klimov, O. (2017). Proximal Policy Optimization Algorithms. arXiv."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1437","DOI":"10.1007\/s11045-018-0611-3","article-title":"Dynamic Time Warping in Classification and Selection of Motion Capture Data","volume":"30","author":"Switonski","year":"2019","journal-title":"Multidimens. Syst. Signal Process."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Oregi, I., Del Campo, I., and L\u00e1zaro, J. (2017, January 18\u201322). On-Line Dynamic Time Warping for Streaming Time Series. Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD), Skopje, Macedonia.","DOI":"10.1007\/978-3-319-71246-8_36"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1186\/s13677-016-0062-z","article-title":"Dynamic Time Warping Multivariate Time Series Similarity Matching Based on Shape Feature and Inclination Angle","volume":"5","author":"Cao","year":"2016","journal-title":"J. Cloud Comput."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"19961","DOI":"10.1038\/s41598-025-02535-5","article-title":"Fitness Exercise Evaluation System Based on Improved DTW Algorithm","volume":"15","author":"Guo","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"13640","DOI":"10.1038\/s41598-025-97637-5","article-title":"Biomechanical Optimization and Reinforcement Learning Provide Insight into Transition from Ankle to Hip Strategy in Human Postural Control","volume":"15","author":"Hong","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_22","unstructured":"Zhou, R., Pan, Y., Chen, J., and Zhang, Z. (2023, January 10\u201316). Natural Actor-Critic for Robust Reinforcement Learning with Function Approximation. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA."},{"key":"ref_23","unstructured":"Yu, Z., and Zhang, X. (2023, January 23\u201329). Actor-Critic Alignment for Offline-to-Online Reinforcement Learning. Proceedings of the International Conference on Machine Learning (ICML), Honolulu, HI, USA."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Lee, S.-S., Choi, J.-H., Byun, J., and Hwang, K.-I. (2025). Dynamic Golf Swing Analysis Framework Based on Efficient Similarity Assessment. Sensors, 25.","DOI":"10.3390\/s25227073"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1080\/00222895.2016.1271297","article-title":"Analysis of Pelvis\u2013Thorax Coordination Patterns of Professional and Amateur Golfers during Golf Swing","volume":"49","author":"Sim","year":"2017","journal-title":"J. Mot. Behav."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"108424","DOI":"10.1016\/j.patcog.2021.108424","article-title":"Velocity-to-Velocity Human Motion Forecasting","volume":"124","author":"Wang","year":"2022","journal-title":"Pattern Recognit."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"G\u00e4rtner, E., Pirinen, A., and Sminchisescu, C. (2020, January 7\u201312). Deep Reinforcement Learning for Active Human Pose Estimation. Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA.","DOI":"10.1609\/aaai.v34i07.6714"},{"key":"ref_28","unstructured":"Nath, S., Verma, R., Ray, A., and Khadilkar, H. (2020). SIBRE: Self-Improvement Based Rewards for Adaptive Feedback in Reinforcement Learning. arXiv."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/2\/392\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,9]],"date-time":"2026-01-09T05:27:24Z","timestamp":1767936444000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/2\/392"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,7]]},"references-count":28,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["s26020392"],"URL":"https:\/\/doi.org\/10.3390\/s26020392","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,7]]}}}