Abstract
The aim of this study was to reveal the clues that humans use to evaluate the impression of dance. Firstly, we investigated whether aesthetic evaluation can be conducted based on movement alone by removing human appearance information, and examining whether stick figure dance evaluations are similar to actual video evaluations. We captured various movements of Japanese dance using motion capture and video cameras, creating two types of videos: actual video and stick figure video. Fifty participants evaluated the aesthetic qualities of the videos based on four categories: proficiency, softness, stability, and beauty. Furthermore, to confirm the importance of perceiving the videos as whole-body movements, we conducted aesthetic evaluation experiments on videos played in reverse and inverted conditions, investigating whether evaluation changes occurred depending on the conditions, similar to those seen in actual videos. The results suggest that even with simplified movements, some consistent evaluation for impression is possible.