Joint correlation analysis of audio-visual dance figures

In this paper we present a framework for analysis of dance figures from audio-visual data. Our audio-visual data is the multiview video of a dancing actor which is acquired using 8 synchronized cameras. The multi-camera motion capture technique of this framework is based on 3D tracking of the markers attached to the dancer's body, using stereo color information. The extracted 31) points are used to calculate the body motion features as 3D displacement vectors. On the other hand, MFC coefficients serve as the audio features. In the first stage of the two stage analysis task, we perform Hidden Markov Model (HMM) based unsupervised temporal segmentation of the audio and body motion features, separately, to extract the recurrent elementary audio and body motion patterns. In the second stage, the correlation of body motion patterns with audio patterns is investigated to create a correlation model that can be used during the synthesis of an audio-driven body animation.