Abstract
The automatic interpretation of dance motion sequences from inertial and pressure-based sensors requires models that maintain strict temporal fidelity, preserve orientation invariance, and produce verifiable reasoning suitable for real-time deployment on constrained hardware. Existing approaches frequently lose accuracy when confronted with heterogeneous choreography, variable sensor placement, or shifting performer-specific kinematics, and they offer limited access to the decision-relevant evidence that underlies their classifications. This study introduces an explainable sensor-driven dance recognition framework constructed around an Adaptive Sensor Normalisation module that performs quaternion-based orientation correction with drift-aware Kalman refinement, a Multi-Scale Motion Feature Extractor that applies tempo-conditioned dilation schedules to capture micro-step transitions and phrase-level rhythmic structures, and a Spatio-Temporal Graph Attention Core that integrates edge-weighted graph convolutions with dual spatial–temporal attention to quantify sensor saliency and temporal concentration. A final Explainable Decision and Feedback Layer links prototype-anchored latent representations with gradient-resolved saliency vectors to expose class-specific motion determinants. The system is optimized for edge-class execution through kernel-level compression and causal attention windows operating on a 512-sample sliding segment. Experiments on three inertial datasets indicate classification accuracy up to 94.2 percent, movement-quality estimation of 92.8 percent, and sub-8.5 millisecond per-frame latency that confirms stability under tempo variation, sensor drift, and partial channel loss.
Similar content being viewed by others
Data availability
The datasets supporting the results of this manuscript are publicly available as follows: ImperialDance dataset: Available at https://github.com/YunZhongNikki/ImperialDance-Dataset?tab=readme-ov-file. CMU-MoCap dataset: Available at https://mocap.cs.cmu.edu/. AIST++ dataset: Available at https://google.github.io/aistplusplus_dataset/factsfigures.html.
References
Li, Y., Yang, G., Su, Z., Li, S. & Wang, Y. Human activity recognition based on multienvironment sensor data. Inf. Fusion 91, 47–63 (2023).
Luptáková, I. D., Kubovčík, M. & Pospíchal, J. Wearable sensor-based human activity recognition with transformer model. Sensors 22, 1911 (2022).
Wang, T. et al. Reslnet: deep residual lstm network with longer input for action recognition. Front. Comp. Sci. 16, 166334 (2022).
Zhao, M. et al. Bidirectional transformer gan for long-term human motion prediction. ACM Trans. Multimed. Comput. Commun. Appl. 19, 1–19 (2023).
Siyao, L. et al. Bailando++: 3d dance gpt with choreographic memory. IEEE Trans. Pattern Anal. Mach. Intell. 45, 14192–14207 (2023).
Chen, D. et al. Two-stream spatio-temporal gcn-transformer networks for skeleton-based action recognition. Sci. Rep. 15, 4982 (2025).
Kishore, P. et al. Machine interpretation of ballet dance: alternating wavelet spatial and channel attention based learning model. IEEE Access (2024).
Baker, B. et al. Computational kinematics of dance: distinguishing hip hop genres. Front. Robotics AI 11, 1295308 (2024).
Ya, L. Research on data collection and posture optimization in dance training using smart wearable devices. Int. J. High Speed Electron. Syst. 2025, 2540492 (2025).
Adalarasu, K., Chetty, R. K., Begum, K. G., Harini, S. & Janardhanan, M. An explainable machine learning (xai) framework for classification of intricate dancing posture among indian bharatanatyam dancers. Appl. Soft Comput. 2025, 112817 (2025).
Li, Y., Peng, G., Du, T., Jiang, L. & Kong, X.-Z. Advancing fractured geothermal system modeling with artificial neural network and bidirectional gated recurrent unit. Appl. Energy 372, 123826 (2024).
Saral, G. B., Bose, K. R. et al. Move match: live dance motion monitoring and feedback system. In 2025 International Conference on Computational, Communication and Information Technology (ICCCIT) 80–85 (IEEE, 2025).
Chen, J., Ye, H., Ying, Z., Sun, Y. & Xu, W. Dynamic trend fusion module for traffic flow prediction. Appl. Soft Comput. 174, 112979 (2025).
Chen, J., Zhang, S. & Xu, W. Scalable prediction of heterogeneous traffic flow with enhanced non-periodic feature modeling. Expert Syst. Appl. 2025, 128847 (2025).
Liang, J. et al. Interaction-aware trajectory prediction for safe motion planning in autonomous driving: a transformer-transfer learning approach. IEEE Trans. Intell. Transp. Syst. 26, 17080–17095. https://doi.org/10.1109/TITS.2025.3588228 (2025).
Zhang, L. et al. Smooth path planning and dynamic contact force regulation for robotic ultrasound scanning. IEEE Robot. Autom. Lett. 10, 10570–10577. https://doi.org/10.1109/LRA.2025.3604746 (2025).
Li, L., Cherouat, A., Snoussi, H. & Wang, T. Grasping with occlusion-aware ally method in complex scenes. IEEE Trans. Autom. Sci. Eng. 22, 5944–5954. https://doi.org/10.1109/TASE.2024.3434610 (2025).
Li, M. Ethnic dance movement recognition based on motion capture sensor and machine learning. Int. J. Inf. Commun. Technol. 25, 81–96 (2024).
Xie, L. Rural folk dance movement recognition based on an improved mcm-svm model in wireless sensing environment. J. Sens. 2023, 9213689 (2023).
Du, L. Design of dance movement recognition algorithm based on 3d motion capture data. In 2023 2nd International Conference on 3D Immersion, Interaction and Multi-sensory Experiences (ICDIIME) 308–312 (IEEE, 2023).
Lin, Z. Dance movement recognition method based on convolutional neural network. In 2023 4th International Conference on Computer Vision, Image and Deep Learning (CVIDL) 255–258 (IEEE, 2023).
Jiang, J. Dance movement recognition technology based on multi feature fusion. In 2023 4th International Conference for Emerging Technology (INCET) 1–5 (IEEE, 2023).
Qu, J. A dance movement quality evaluation model using transformer encoder and convolutional neural network. Sci. Rep. 14, 32058 (2024).
Zhen, N. & Jiang, Y. Dance video action recognition algorithm based on improved hypergraph convolutional networks. Syst. Soft Comput. 2025, 200247 (2025).
Ning, T., Zhang, T. & Huang, G. Fctnet: fusion of 3d cnn and transformer dance action recognition network. J. Intell. Fuzzy Syst. 48, 23–31 (2025).
Li, H. & Huang, X. Intelligent dance motion evaluation: an evaluation method based on keyframe acquisition according to musical beat features. Sensors 24, 6278 (2024).
Zheng, D. & Yuan, Y. Pose recognition of dancing images using fuzzy deep learning technique in an iot environment. Expert. Syst. 42, e13422 (2025).
Wang, X. Prediction algorithm of dance movement rhythm and coordination based on time series analysis. In 2024 International Conference on Power, Electrical Engineering, Electronics and Control (PEEEC) 556–561 (IEEE, 2024).
Jiao, Y. Optimizing dance training programs using deep learning: Exploring motion feedback mechanisms based on pose recognition and prediction. Int. J. Adv. Comput. Sci. Appl. 15, 256 (2024).
Chen, J. Motion decomposition and guidance technology in dance teaching based on motion feedback system. In 2023 International Conference on Intelligent Computing, Communication & Convergence (ICI3C) 162–166 (IEEE, 2023).
Cai, X., Lu, Q., Li, F., Liu, S. & Hu, Y. Hand movement recognition and analysis based on deep learning in classical hand dance videos. In Computer Graphics International Conference 53–64 (Springer, 2023).
Zhong, Y. & Demiris, Y. Dancemvp: Self-supervised learning for multi-task primitive-based dance performance assessment via transformer text prompting. In Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38 10270–10278 (2024).
CMU Graphics Lab. Cmu graphics lab motion capture database. http://mocap.cs.cmu.edu/ (2003). Funded by NSF Grant EIA-0196217.
Li, R., Yang, S., Ross, D. A. & Kanazawa, A. Ai choreographer: music conditioned 3d dance generation with aist++. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) 13401–13412 (2021).
Funding
This research was funded by special research topic of cultural exchange of Ministry of Education (Grant Number CCIPE-YXSJ-20240060), key topics of open online course guidance for undergraduate universities in Guangdong Province (Grant Number 2022ZXKC361), and Guangzhou Musicians Association “Music Culture research” and “Primary and secondary school music education reform project” (Grant Number 24GZYX003).
Author information
Authors and Affiliations
Contributions
Jinying Han contributed to the design of the framework, the development of the methodology, and the writing of the J.H. designed the framework, developed the methodological components, performed the data analysis, and prepared the main manuscript text. Shan-Wang-S.W. assisted with supervision, coordinated data collection, prepared the experimental setup, implemented the X-DanceNet framework, reviewed, and edited the manuscript. All authors reviewed and edited the manuscript. Jiayin Gao contributed to the conceptualization and conceptual development of the study, supported the theoretical formulation, and reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Han, J., Wang, S. & Gao, J. An explainable real time sensor graph transformer for dance recognition. Sci Rep (2026). https://doi.org/10.1038/s41598-025-34691-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-025-34691-z
