Synthesizing novel animations of periodic dances

Introduction

We present an efficient algorithm for synthesizing novel, arbitrarily long animations of periodic dances. The input to the proposed method is motion capture data acquired from markeless visual observations of a human performing a periodic dance. Several experiments have been conducted with real recordings of Greek folk dances.

The obtained results are very promising and indicate the efficacy of the proposed approach, as well as its tolerance to dynamic and noisy human motion capture input.

Methodology

Figure 2: Results of the proposed methods [1]-[2].

  • The first module realizes the temporal segmentation of the complex motion capture data into motion patterns.

  • Next, using the extracted motion patterns, we construct a motion graph and we employ an efficient algorithm for graph exploration that yields an unseen animation [2].

  • Finally, in the case that a music is given, a beat synchronous animation is created by applying the method proposed in [1] (see Figs 1,2).

The proposed method improves the motion signal continuity during motion patterns transitions and reduces the noise of the original motion signal. Moreover, the proposed method is able to synthesize realistic animations with high motion variability even in cases where the number of motion patterns is low.

Downloads

    • You can download video results containing ten beat synchronous dance animation videos of the proposed method [1] .

    • You can download video results containing several synthetic dance animation videos of the proposed method [2].

    • See the corresponding readme.txt files and [1-2] for more details.

Related Publications

[1] C. Panagiotakis, Andre Holzapfel, Damien Michel, and Antonis Argyros, Beat Synchronous Dance Animation based on Periodic Motion and Music Tempo Analysis, International Symposium on Visual Computing, 2013..

[2] C. Panagiotakis, Damien Michel, and Antonis Argyros,Temporal segmentation and seamless stitching of motion patterns for synthesizing novel animations of periodic dances, International Conference on Pattern Recognition, 2014.

ACKNOWLEDGMENTS

This research has been co-financed by the European Union (European Social Fund - ESF) and Greek national funds through the Operational Program “Education and Life- long Learning” of the National Strategic Reference Frame- work (NSRF) - Research Funding Project: THALIS-UOA-ERASITECHNIS MIS 375435.