Application of Classification Method of Emotional Expression Type Based on Laban Movement Analysis to Design Creation

摘要

Emotion estimation is one of the most essential research areas along with the progress of human sensing and AI technologies. We have already proposed a classification method of emotional expression type based on Laban movement analysis, which is a typical theory for dancers. In this study, we applied the classification method to design creation, which is typically performed in digital fabrication. First, we made clear what kinds of emotions are evoked in digital fabrication tasks by using the evaluation grid method, and we analyzed the emotions by constructing a core affect model for the task. Next, we performed an experiment to measure the dataset of body motions and emotions by performing an experiment using SONY FES Watch U. By using the dataset, we classified users by body motions, estimated the evoked emotions by using the classified dataset. and realized emotion estimation at about 80%. We could estimate emotions even when the body motions were not so large or activated compared with the fabrication task. The results showed the general effectiveness of the classification method.