Performance monitoring and evaluation in dance teaching with mobile sensing technology

One of the most important jobs of dance teacher is to detect and estimate the performances of the students during their practices, including confirming the completion of the moves in and out of class as well as evaluating the performance . However, it would be a heavy burden for teachers who do it by themselves, and traditional assist of video techniques is quite expensive and inconvenient for ordinary learners. Therefore, in this paper, we apply sensors within the smart phone to follow the performance of dance trainers and estimate the correctness of motion gesture and rhythm management by measuring the similarity between the motion data of trainers and the standard data with a dynamic time warping based algorithm. And we propose an automatic grading system to assess users' performances, including the comprehensive evaluation grade as well as the rhythm situation in each part. We apply our solution in the real world practice and compare its results with those that given by experts. According to the results of our experiments, this scheme is able to judge the dancing practice quite accurately and is accepted and confirmed by teachers and experts in dance field. Hopefully, it will be extended to other sports fields as well, such as aerobics and yoga.