SYNCUP: Vision-Based Practice Support for Synchronized Dancing

Abstract

The popularization of smart mobile devices has changed how people practice dancing: dancers typically use smartphones to take video records while practicing. However, to review their videos, they primarily use simple video player interfaces, which do not provide sufficient support for dancers. In this work, we present SyncUp, a system that provides multiple interactive visualizations to support the practice of synchronized dancing and liberate users from manual inspection of recorded practice videos. SyncUp helps dancers quickly identify poorly synchronized body parts and dance portions through our interactive visualization designs and dance performance quantification methods. Our extensive evaluation study shows that SyncUp predictions are highly correlated with human rating scores, and reveal practical benefits of SyncUp to dancers.