An automated evaluator for a classical dance — Bharatanatyam (Nritta)

This project implements a system that automatically evaluates a dance performance against a gold standard and provides a scorecard to the performer in the form of a graph with suggestions for improvement. The system acquires the motion of the performer via joint detection. A camera is used to capture the front view of the dancer. The video is then fed to the system that breaks it into frames. Eight important joints of the body are detected using the HAAR classification technique in each frame and the changes in position of the joints are used to find the joint velocity. A scoring algorithm is devised which grades a dancer by comparison against a gold standard. Scoring is done separately for upper body and lower body movements. A final score is provided which is a combination of the upper body and lower body scores. Apart from scoring, the system also tries to identify areas the dancer can improve. This system tries to eliminate human effort required in evaluating a dancer during the examinations by automating the entire process.