Confidence-Adaptive AI-Instructor Feedback Fusion for Enhanced Engagement in Online Latin Dance Learning
DOI:
https://doi.org/10.18533/gekxh943Keywords:
Dance education, Adaptive feedback, Human-AI collaboration, Pose estimation, Reinforcement learningAbstract
Online dance training has significant challenges in providing personalized feedback, particularly in culturally nuanced styles such as Latin dance. The paper introduces a confidence-adaptive feedback system that integrates AI-generated corrections with real-time input from human instructors to enhance the accuracy and engagement of the learning process. The method uses real-time pose estimation to figure up a composite confidence score by combining how accurate the movement is with how hard the person thinks it is. A motion analyzer based on Transformers can pick up on changes in rhythm, posture, and coordination. A multi-armed bandit system improves feedback allocation by putting more weight on human instruction when confidence is low. The system enhanced movement accuracy by 40% and practice repetitions by 19.1% compared to systems that just used AI. This was shown in a 12-week study with 120 participants. This mixed approach gives online dance teachers a flexible plan that works with people from many different cultures. It finds a balance between technical accuracy and personal intuition.
References
Baker, R. (2016). Using learning analytics in personalized learning. In Handbook on Personalized Learning for States, Districts, and Schools.
Cao, Z., Hidalgo, G., Simon, T., Wei, S. E., & Sheikh, Y. (2019). OpenPose: Realtime multi-person 2D pose estimation using part affinity fields. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(1), 172–186.
Chen, J. (2024). Dance education using digital technologies: Enhancing effectiveness by facilitating student-teacher feedback. Theatre, Dance and Performance Training, 15(2), 123–135.
El Raheb, K., Kasomoulis, A., Katifori, A., & Ioannidis, Y. (2018). A web-based system for annotation of dance multimodal recordings by dance practitioners and experts. Proceedings of the 5th International Conference on Movement and Computing, 1–8.
Ferguson, S., Schubert, E., & Stevens, C. J. (2014). Dynamic dance warping: Using dynamic time warping to compare dance movement performed under different conditions. Proceedings of the International Symposium on Performance Science, 247–252.
Halder, A., & Tayade, A. (2021). Real-time vernacular sign language recognition using MediaPipe and machine learning. International Journal of Progressive Research and Reviews, 8(3), 45–52.
Hutson, J., & Plate, D. (2023). Human-AI collaboration for smart education: Reframing applied learning to support metacognition. IntechOpen. https://doi.org/10.5772/intechopen.108765
Lewis, F. L., & Vrabie, D. (2009). Reinforcement learning and adaptive dynamic programming for feedback control. IEEE Circuits and Systems Magazine, 9(3), 32–50.
Li, M., Miao, Z., & Ma, C. (2019). Dance movement learning for labanotation generation based on motion-captured data. IEEE Access, 7, 123456–123465.
Li, R., Zhao, J., Zhang, Y., Su, M., Ren, Z., & Wang, H. (2023). FineDance: A fine-grained choreography dataset for 3D full body dance generation. Proceedings of the IEEE/CVF International Conference on Computer Vision, 4567–4576.
Ma, R., Kiyasseh, D., Laca, J. A., Kocielnik, R., & Anandkumar, A. (2024). Artificial intelligence-based video feedback to improve novice performance on robotic suturing skills: A pilot study. Journal of Endourology, 38(4), 321–329.
Martínez-González, A., Villamizar, M., & Odobez, J.-M. (2021). Pose transformers (POTR): Human motion prediction with non-autoregressive transformers. IEEE/CVF International Conference on Computer Vision Workshops, 1234–1243.
Mui, J., Lin, F., & Dewan, M. A. A. (2021). Multi-armed bandit algorithms for adaptive learning: A survey. International Conference on Artificial Intelligence and Soft Computing, 789–798.
Munaro, M., Basso, A., Fossati, A., & Menegatti, E. (2014). 3D reconstruction of freely moving persons for re-identification with a depth sensor. International Conference on Robotics and Automation, 4512–4519.
Raheb, K. E., Stergiou, M., Katifori, A., & Ioannidis, Y. (2019). Dance interactive learning systems: A study on interaction workflow and teaching approaches. ACM Computing Surveys, 52(3), 1–37.
Rothmund, I. V. (2023). Student-centred learning and dance technique: BA students’ experiences of learning in contemporary dance. Research in Dance Education, 24(1), 45–58.
Schlagenhauf, F., Sreeram, S., & Singh, S. (2018). Comparison of Kinect and Vicon motion capture of upper-body joint angle tracking. 2018 IEEE 14th International Conference on Control, Automation, Robotics and Vision, 1–6.
Sun, G., Wong, Y., Cheng, Z., & Kankanhalli, M. (2020). DeepDance: Music-to-dance motion choreography with adversarial learning. IEEE Transactions on Multimedia, 22(5), 1234–1245.
Sun, K., Xiao, B., Liu, D., & Wang, J. (2019). Deep high-resolution representation learning for human pose estimation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 5693–5703.
Tuomi, I. (2023). A framework for socio-developmental ethics in educational AI. AIS Electronic Library (AISeL). https://aisel.aisnet.org/amcis2023/education/5
Yang, A. C. M., Chen, I. Y. L., Flanagan, B., & Ogata, H. (2022). How students’ self-assessment behavior affects their online learning performance. Computers and Education Open, 3, 100089.
Yu, J., Yu, S., & Chen, L. (2025). Using hybrid intelligence to enhance peer feedback for promoting teacher reflection in video-based online learning. British Journal of Educational Technology, 56(2), 234–245.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Zheng Lin, Dr. Leng Poh Gee
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
