Dancing with Turks

Authors:

I-Kao Chiang Dept. of Computer and Information Science University of Pennsylvania Philadelphia, PA, USA	Ian Spiro Dept. of Computer Science New York University, New York, NY, USA	Seungkyu Lee Dept. of Computer Engineering KyungHee University, Yongin-si,South Korea
Alyssa Lees Dept. of Computer Science New York University, New York, NY, USA	Jingchen Liu School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park, PA, USA	Chris Bregler Dept. of Computer Science New York University, New York, NY, USA
Yanxi Liu School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park, PA, USA

Abstract:

Dance is a dynamic art form that reflects a wide range of cultural diversity and individuality. With the advancement of motion capture technology in combination with crowd-sourcing and machine learning algorithms, we explore the complex relationship between perceived dance quality/dancer gender, and dance movements/music. As a feasibility study, we construct a computational framework for an analysis-synthesis-feedback loop using a novel and synchronized multimedia dance-music texture representation. Furthermore, we integrate crowd sourcing, music and motion-capture data, and machine learning-based methods for dance segmentation, analysis and synthesis of new dancers. A quantitative validation of this framework on a mocap dataset of 172 dancers evaluated by more than 400 independent on-line raters demonstrates significant correlation between human perception and the algorithmically intended dance quality or gender of synthesized dancers. The technology illustrated in this work has a high potential to advance the multimedia entertainment industry via dancing with Turks.

Videos:


Dance Party!

Ability Control

Dancing to New Music

Citation:

Dancing with Turks
I-Kao Chiang, Ian Spiro, Seungkyu Lee, Alyssa Lees, Jingchen Liu, Chris Bregler, Yanxi Liu
long paper (10 pages), in press, ACM Multimedia 2015

Full Paper PDF

Mechanical Turker Comments:

Generated Word Cloud from User Data using Amuller's python implementation

Acknowledgment:

We thank Professors Robert Trivers and Lee Cronk of Rutgers University for sharing their motion captured data of Jamaican teenager dancers.
Brian VanLeeuwen contributed to Figure 3(A). I-Kao Chiang (first author) worked on part of this project for his BS honors thesis while at PSU. This work is supported in part by NSF grants IIS-1248076 and IIS-1144938.