Rhythm is a Dancer: Music-Driven Motion Synthesis With Global Structure

被引:22
作者
Aristidou, Andreas [1 ,2 ]
Yiannakidis, Anastasios [1 ,2 ]
Aberman, Kfir [3 ]
Cohen-Or, Daniel [3 ]
Shamir, Ariel [4 ]
Chrysanthou, Yiorgos [1 ,2 ]
机构
[1] Univ Cyprus, Dept Comp Sci, CY-1678 Nicosia, Cyprus
[2] CYENS Ctr Excellence, CY-1016 Nicosia, Cyprus
[3] Tel Aviv Univ, Dept Comp Sci, IL-6997801 Tel Aviv, Israel
[4] Reichman Univ, Interdisciplinary Ctr Herzliya, Dept Comp Sci, IL-4610101 Herzliyya, Israel
关键词
Animation; global structure consistency; motion motifs; music-driven; motion signatures; BEAT TRACKING; STYLE; PHYSICS;
D O I
10.1109/TVCG.2022.3163676
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Synthesizing human motion with a global structure, such as a choreography, is a challenging task. Existing methods tend to concentrate on local smooth pose transitions and neglect the global context or the theme of the motion. In this work, we present a music-driven motion synthesis framework that generates long-term sequences of human motions which are synchronized with the input beats, and jointly form a global structure that respects a specific dance genre. In addition, our framework enables generation of diverse motions that are controlled by the content of the music, and not only by the beat. Our music-driven dance synthesis framework is a hierarchical system that consists of three levels: pose, motif, and choreography. The pose level consists of an LSTM component that generates temporally coherent sequences of poses. The motif level guides sets of consecutive poses to form a movement that belongs to a specific distribution using a novel motion perceptual-loss. And the choreography level selects the order of the performed movements and drives the system to follow the global structure of a dance genre. Our results demonstrate the effectiveness of our music-driven framework to generate natural and consistent movements on various dance types, having control over the content of the synthesized motions, and respecting the overall structure of the dance.
引用
收藏
页码:3519 / 3534
页数:16
相关论文
共 92 条
  • [1] Generative Autoregressive Networks for 3D Dancing Move Synthesis From Music
    Ahn, Hyemin
    Kim, Jaehun
    Kim, Kihyun
    Oh, Songhwai
    [J]. IEEE ROBOTICS AND AUTOMATION LETTERS, 2020, 5 (02): : 3501 - 3508
  • [2] Alemi O., 2017, PROC 23 ACM SIGKDD I, P6
  • [3] Style-Controllable Speech-Driven Gesture Synthesis Using Normalising FlowsKeywords
    Alexanderson, Simon
    Henter, Gustav Eje
    Kucherenko, Taras
    Beskow, Jonas
    [J]. COMPUTER GRAPHICS FORUM, 2020, 39 (02) : 487 - 496
  • [4] Look, Listen and Learn
    Arandjelovic, Relja
    Zisserman, Andrew
    [J]. 2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, : 609 - 617
  • [5] Digital Dance Ethnography: Organizing Large Dance Collections
    Aristidou, Andreas
    Shamir, Ariel
    Chrysanthou, Yiorgos
    [J]. ACM JOURNAL ON COMPUTING AND CULTURAL HERITAGE, 2019, 12 (04):
  • [6] Deep Motifs and Motion Signatures
    Aristidou, Andreas
    Cohen-Or, Daniel
    Hodgins, Jessica K.
    Chrysanthou, Yiorgos
    Shamir, Ariel
    [J]. ACM TRANSACTIONS ON GRAPHICS, 2018, 37 (06):
  • [7] Style-based motion analysis for dance composition
    Aristidou, Andreas
    Stavrakis, Efstathios
    Papaefthimiou, Margarita
    Papagiannakis, George
    Chrysanthou, Yiorgos
    [J]. VISUAL COMPUTER, 2018, 34 (12) : 1725 - 1737
  • [8] Beaudoin Philippe, 2008, P ACM SIGGRAPH EUR S, P117
  • [9] Dance to the beat: Synchronizing motion to audio
    Bellini R.
    Kleiman Y.
    Cohen-Or D.
    [J]. Computational Visual Media, 2018, 4 (3) : 197 - 208
  • [10] A tutorial on onset detection in music signals
    Bello, JP
    Daudet, L
    Abdallah, S
    Duxbury, C
    Davies, M
    Sandler, MB
    [J]. IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 2005, 13 (05): : 1035 - 1047