Mining crucial features for automatic rehabilitation coaching systems

Our goal is to develop a system for coaching human motions (e.g. rehabilitation). Such a coaching system should have several function such as motion measurement, evaluation, and feedback. Among all, this paper focuses on how to modify a user's motion so that it gets closer to the good template of a target motion. To this end, it is important to efficiently advise the user to emulate the crucial features that define the good template. The proposed method automatically mines the crucial features of any kind of motions from a set of all motion features. The crucial features are mined based on feature sparsification through binary classification between the samples of good and other motions.