Producing "Open-Style" Choreography for K-Pop Music with Deep Learning

Songha  Ban; Sharon Ong

Abstract

The goal of this thesis to generate natural and beat-matching choreography from music using deep learning. We compare different pose estimators to create a dataset of human figures for dance generation. Our deep learning framework comprises of a music encoder to create music features which is fed to a pose generator to create dance pose sequences and a music feature generator which reconstructed music features from output poses to improve music feature encoding. Our results showed a pose estimator with a GRU music encoder, generated more natural dance movements which match K-Pop music compared to previous work.

Original language	English
Publication status	Published - 11 Nov 2021
Event	33rd Benelux Conference on Artificial Intelligence and the 30th Belgian Dutch Conference on Machine Learning - University of Luxembourg, Luxembourg, Luxembourg Duration: 10 Nov 2021 → 12 Nov 2021 https://bnaic2021.uni.lu/

Conference

Conference	33rd Benelux Conference on Artificial Intelligence and the 30th Belgian Dutch Conference on Machine Learning
Abbreviated title	BNAIC/BENELEARN 2021
Country/Territory	Luxembourg
City	Luxembourg
Period	10/11/21 → 12/11/21
Internet address	https://bnaic2021.uni.lu/

Cite this

@conference{ad8f956695004de2ac866c46288520b1,

title = "Producing {"}Open-Style{"} Choreography for K-Pop Music with Deep Learning",

abstract = "The goal of this thesis to generate natural and beat-matching choreography from music using deep learning. We compare different pose estimators to create a dataset of human figures for dance generation. Our deep learning framework comprises of a music encoder to create music features which is fed to a pose generator to create dance pose sequences and a music feature generator which reconstructed music features from output poses to improve music feature encoding. Our results showed a pose estimator with a GRU music encoder, generated more natural dance movements which match K-Pop music compared to previous work.",

author = "Songha Ban and Sharon Ong",

year = "2021",

month = nov,

day = "11",

language = "English",

note = "33rd Benelux Conference on Artificial Intelligence and the 30th Belgian Dutch Conference on Machine Learning , BNAIC/BENELEARN 2021 ; Conference date: 10-11-2021 Through 12-11-2021",

url = "https://bnaic2021.uni.lu/",

}

TY - CONF

T1 - Producing "Open-Style" Choreography for K-Pop Music with Deep Learning

AU - Ban, Songha

AU - Ong, Sharon

PY - 2021/11/11

Y1 - 2021/11/11

N2 - The goal of this thesis to generate natural and beat-matching choreography from music using deep learning. We compare different pose estimators to create a dataset of human figures for dance generation. Our deep learning framework comprises of a music encoder to create music features which is fed to a pose generator to create dance pose sequences and a music feature generator which reconstructed music features from output poses to improve music feature encoding. Our results showed a pose estimator with a GRU music encoder, generated more natural dance movements which match K-Pop music compared to previous work.

AB - The goal of this thesis to generate natural and beat-matching choreography from music using deep learning. We compare different pose estimators to create a dataset of human figures for dance generation. Our deep learning framework comprises of a music encoder to create music features which is fed to a pose generator to create dance pose sequences and a music feature generator which reconstructed music features from output poses to improve music feature encoding. Our results showed a pose estimator with a GRU music encoder, generated more natural dance movements which match K-Pop music compared to previous work.

M3 - Abstract

T2 - 33rd Benelux Conference on Artificial Intelligence and the 30th Belgian Dutch Conference on Machine Learning

Y2 - 10 November 2021 through 12 November 2021

ER -

Producing "Open-Style" Choreography for K-Pop Music with Deep Learning

Abstract

Conference

Fingerprint

Cite this