Dance motion analysis and editing using hilbert-huang transform

Author

Ran Dong, Dongsheng Cai, and Nobuyoshi Asai

Abstract

Human motions (especially, dance motions) are very noisy and it is difficult to analyze the motions. To resolve this problem, we propose a new method to decompose and edit the motions using the Hilbert-Huang transform (HHT). The HHT decomposes a chromatic signal into "monochromatic" signals that are the so-called Intrinsic Mode Functions (IMFs) using an Empirical Mode Decomposition (EMD)Huang 2014. The HHT has the advantage to analyze non-stationary and nonlinear signals like human joint motions over the FFT or Wavelet transform. In the present research, we propose a new framework to analyze a famous Japanese threesome pop singer group "Perfume". Then using the NA-MEMD, we decompose dance motions into motion (choreographic) primitives or IMFs, which can be scaled, combined, subtracted, exchanged, and modified self-consistently.

Source

ACM SIGGRAPH 2017 Talks (SIGGRAPH '17)

Comments

Hilbert-Huang transform (HHT)

URL

Link

Tag

#SIGGRAPH #2017

#motion #motion_analysis #motion_editing #motion_blending #motion_synthesis

#解析/Analysis