This master's thesis examines the development and evaluation of an educational application for learning Slovenian folk dances using virtual reality technology. Currently, affordable VR headsets available on the market do not support full-body tracking. However, learning dance in virtual reality requires full-body tracking. We address this limitation by integrating a low-cost external camera and a machine learning-based motion capture model, which ensures accurate real-time body tracking.
In collaboration with experienced folk dancers, we recorded over 60 dances that form the foundation for learning. The application is developed in the Unity environment and includes several Slovenian folk dances with modular learning scenarios, allowing dance practice either as exercise or in evaluation mode, which provides feedback.
For the evaluation, participants were divided into two groups. The first group, the video group, learned dances from recordings, while the second group, the VR group, used virtual reality headsets for learning. The comparison between the groups provided results for assessing whether virtual reality is suitable for learning folk dances and what its advantages and disadvantages are in this context. Evaluation by folk dance experts added further value to the comparison, as they assessed from a professional perspective the impact of virtual reality on learning and participant engagement.
The result is an innovative application for learning folk dances that does not require expensive additional equipment to track the VR headset user's body. At the same time, it offers a solution for the digital preservation and dissemination of Slovenian cultural heritage, as it combines affordable technology with satisfactory motion capture quality and provides the opportunity to learn dance in a home environment.
|
