Enhancing the Educational Potential of Online Movement Videos: System Development and Empirical Studies with TikTok Dance Challenges

Research Background and Issues

• What problems or challenges did the authors identify?
  The authors pointed out that while online motion videos (e.g., TikTok dance tutorial videos) provide rich content, their educational potential is not fully realized in standard video formats. These videos lack structured guidance, making it difficult for beginners to effectively learn complex movements. Additionally, existing dance teaching systems often rely on expensive equipment or manual content creation, limiting scalability and adaptability.

• Why is this issue important?
  With the growth of online learning, effectively teaching motor skills through videos is a crucial area. Providing scalable automated dance teaching tools can benefit resource-limited individuals eager to learn.

• Research Motivation and Related Work
  TikTok dance challenges, combining social interaction and skill development, offer an ideal model for studying effective online learning environments. However, there is insufficient research on the intrinsic educational mechanisms of TikTok tutorial videos. The authors aim to address this gap by developing an automated teaching system and evaluating its educational potential.

Solution

• What methods or solutions did the authors propose?
  The authors developed a platform capable of automatically generating structured practice plans from TikTok dance videos. The plans are based on incremental part-learning and fading guidance principles and are presented through a web-based interactive interface, equipped with visual aids such as skeleton overlays and static motion diagrams.

• What are the innovative aspects of this solution?

  1. Automated Practice Plan Generation: The system uses image processing techniques to extract skeleton data from original dance videos, analyze them, and construct segmented learning plans.
  2. Integration of Technology and Educational Theory: The teaching design is optimized based on motor learning theories, incorporating structured part-learning and gradual fading guidance.
  3. Lightweight Equipment: The system supports common devices (e.g., cameras and web interfaces), reducing hardware requirements and enhancing scalability.

• What are the implementation steps and key technologies used?

  1. MediaPipe is used to extract skeleton joint data from dancers in the videos.
  2. Dance movements (e.g., keyframes and music beats) are analyzed to automatically generate segmented learning plans.
  3. An interactive interface is designed with step-by-step guided practice processes, including demonstration, practice, testing, and integration phases.
  4. Visual aids (e.g., skeleton overlays and static motion diagrams) are provided to help learners imitate movements and master rhythms.

Research Outcomes

• What specific outcomes were achieved?

  1. Improved User Learning Outcomes: Compared to TikTok tutorial videos, the system significantly enhanced learners' movement accuracy and learning experience, particularly for complex dances.
  2. Positive User Feedback: Users generally found the system's slow practice feature, segmented learning, and skeleton overlay aids very helpful.

• What advantages does it have compared to existing solutions?
  Compared to traditional TikTok videos or expensive manual dance teaching systems, the platform's automated generation and theory-supported design significantly improve user learning efficiency while offering greater scalability and device accessibility.

• What are the experimental or evaluation results?
  Two user studies were conducted to evaluate the system:

  1. Study 1: Skeleton overlay aids significantly improved learning outcomes compared to original TikTok videos, and segmented learning methods proved crucial.
  2. Study 2: While segmentation and emoji annotations in TikTok tutorials increased user preference, these features alone did not significantly enhance learning outcomes.

• Limitations and Future Directions

  1. Limitations:
     • TikTok dance content is short and narrow in format, making it difficult to represent more complex dance styles.
     • Emoji annotations and segmented learning used independently did not achieve optimal results, indicating the need for more structured integration.
  2. Future Research Directions:
     • Explore deeper integration of segmented learning and annotations into teaching systems.
     • Extend the system to more complex dance styles and other domains of motor skill learning.
     • Investigate the acceptance and learning outcomes of the system across different age groups and backgrounds.

Conclusion

This study developed an innovative system that automatically generates dance teaching plans using TikTok videos, significantly improving user learning experience and outcomes. It also highlights the importance of segmented learning and visual aids in motor skill acquisition. Future research should further optimize these features and expand the system's application scope.