AI-Enhanced Assistive Tools with Hand Skeleton Analysis in Online Arts Education

The global COVID-19 pandemic has significantly impacted education worldwide, leading to a transition from traditional in-person teaching to asynchronous online learning. Thanks to the valuable contributions of numerous high-quality educational content creators, online audiovisual materials have opened up boundless opportunities for cross-border, crossdisciplinary online learning. This removes geographical and temporal barriers and broadens the spectrum of learning resources available to individuals with diverse needs. Particularly for individuals with disabilities, mobility impairments, or social anxiety, online learning offers an inclusive and accessible environment, fostering the democratization and equity of education and ensuring highquality education regardless of physical conditions or locations. However, unlike foundational subjects, the learning process for arts heavily relies on live demonstrations by teachers and imitation by students, presenting a challenge in the online learning setting especially true for beginners as learning arts often necessitates personalized guidance and real-time feedback. To address this challenge, Mediapipe technology was used to capture the hand movements of both teachers and students in this study. Aeep learning model was used to analyze the similarity of pen-holding postures between students and teachers, thereby providing real-time feedback to supplement the lack of interactivity in online drawing learning and enhance learning outcomes. The developed learning method simulates the role of teachers in traditional classrooms and widens the scope of arts education, providing AI-powered tools for those interested in cultivating aesthetics and lifelong learning opportunities. For the care of the elderly, in addition to enhancing emotional and psychological well-being through aesthetics, the method can be utilized for rehabilitation or the assessment of degenerative conditions.