Title :
Rhythmic motion analysis using motion capture and musical information
Author :
Shiratori, Takaaki ; Nakazawa, Atsushi ; Ikeuchi, Katsushi
Author_Institution :
Dept. of Inf. & Commun. Eng., Univ. of Tokyo, Japan
fDate :
30 July-1 Aug. 2003
Abstract :
The number of Japanese traditional dancers is decreasing. Without performers, some dances will disappear because conventional media such as paper cannot record them. We have proposed an archiving method specifically for dancing patterns. Our method has four main stages: 1) digitizing motions by motion capture systems; 2) analyzing motions; 3) synthesizing motions; and 4) reproducing the dance motions by CG (computer-generated) and humanoid robots. In order to effectively record the movement patterns, motion primitives are extracted. Each motion primitive describes each basic motion. However, most previous primitive extraction failed to provide the rhythm, which results in unrhythmic, synthesized motion. In this paper, we propose a motion analysis method, which integrates music rhythm into the motion primitives. Our experiment confirmed that our motion analysis yielded motion primitives in accordance to the music rhythm.
Keywords :
computer animation; image motion analysis; image retrieval; image sequences; mobile robots; music; Japanese traditional dance; computer-generated robot; dance motion reproduction; dancing pattern; humanoid robot; motion capture system; motion digitization; motion primitive; motion synthesis; movement pattern recording; music rhythm; musical information; rhythmic motion analysis; Character generation; Communication industry; Computer graphics; Filter bank; Humanoid robots; Humans; Interpolation; Motion analysis; Pattern analysis; Rhythm;
Conference_Titel :
Multisensor Fusion and Integration for Intelligent Systems, MFI2003. Proceedings of IEEE International Conference on
Print_ISBN :
0-7803-7987-X
DOI :
10.1109/MFI-2003.2003.1232638