Abstract
Synthesizing human's movements such as dancing is a flourishing research field which has several applications in computer graphics. Recent studies have demonstrated the advantages of deep neural networks (DNNs) for achieving remarkable performance in motion and music tasks with little effort for feature pre-processing. However, applying DNNs for generating dance to a piece of music is nevertheless challenging, because of 1) DNNs need to generate large sequences while mapping the music input, 2) the DNN needs to constraint the motion beat to the music, and 3) DNNs require a considerable amount of hand-crafted data. In this study, we propose a weakly supervised deep recurrent method for real-time basic dance generation with audio power spectrum as input. The proposed model employs convolutional layers and a multilayered Long Short-Term memory (LSTM) to process the audio input. Then, another deep LSTM layer decodes the target dance sequence. Notably, this end-to-end approach has 1) an auto-conditioned decode configuration that reduces accumulation of feedback error of large dance sequence, 2) uses a contrastive cost function to regulate the mapping between the music and motion beat, and 3) trains with weak labels generated from the motion beat, reducing the amount of hand-crafted data. We evaluate the proposed network based on i) the similarities between generated and the baseline dancer motion with a cross entropy measure for large dance sequences, and ii) accurate timing between the music and motion beat with an F-measure. Experimental results revealed that, after training using a small dataset, the model generates basic dance steps with low cross entropy and maintains an F-measure score similar to that of a baseline dancer.
| Original language | English |
|---|---|
| Title of host publication | 2019 International Joint Conference on Neural Networks, IJCNN 2019 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9781728119854 |
| DOIs | |
| Publication status | Published - 2019 Jul |
| Event | 2019 International Joint Conference on Neural Networks, IJCNN 2019 - Budapest, Hungary Duration: 2019 Jul 14 → 2019 Jul 19 |
Publication series
| Name | Proceedings of the International Joint Conference on Neural Networks |
|---|---|
| Volume | 2019-July |
Conference
| Conference | 2019 International Joint Conference on Neural Networks, IJCNN 2019 |
|---|---|
| Country | Hungary |
| City | Budapest |
| Period | 19/7/14 → 19/7/19 |
Fingerprint
Keywords
- Contrastive loss
- Dance generation
- Deep recurrent networks
ASJC Scopus subject areas
- Software
- Artificial Intelligence
Cite this
Weakly-Supervised Deep Recurrent Neural Networks for Basic Dance Step Generation. / Yalta, Nelson; Watanabe, Shinji; Nakadai, Kazuhiro; Ogata, Tetsuya.
2019 International Joint Conference on Neural Networks, IJCNN 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8851872 (Proceedings of the International Joint Conference on Neural Networks; Vol. 2019-July).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Weakly-Supervised Deep Recurrent Neural Networks for Basic Dance Step Generation
AU - Yalta, Nelson
AU - Watanabe, Shinji
AU - Nakadai, Kazuhiro
AU - Ogata, Tetsuya
PY - 2019/7
Y1 - 2019/7
N2 - Synthesizing human's movements such as dancing is a flourishing research field which has several applications in computer graphics. Recent studies have demonstrated the advantages of deep neural networks (DNNs) for achieving remarkable performance in motion and music tasks with little effort for feature pre-processing. However, applying DNNs for generating dance to a piece of music is nevertheless challenging, because of 1) DNNs need to generate large sequences while mapping the music input, 2) the DNN needs to constraint the motion beat to the music, and 3) DNNs require a considerable amount of hand-crafted data. In this study, we propose a weakly supervised deep recurrent method for real-time basic dance generation with audio power spectrum as input. The proposed model employs convolutional layers and a multilayered Long Short-Term memory (LSTM) to process the audio input. Then, another deep LSTM layer decodes the target dance sequence. Notably, this end-to-end approach has 1) an auto-conditioned decode configuration that reduces accumulation of feedback error of large dance sequence, 2) uses a contrastive cost function to regulate the mapping between the music and motion beat, and 3) trains with weak labels generated from the motion beat, reducing the amount of hand-crafted data. We evaluate the proposed network based on i) the similarities between generated and the baseline dancer motion with a cross entropy measure for large dance sequences, and ii) accurate timing between the music and motion beat with an F-measure. Experimental results revealed that, after training using a small dataset, the model generates basic dance steps with low cross entropy and maintains an F-measure score similar to that of a baseline dancer.
AB - Synthesizing human's movements such as dancing is a flourishing research field which has several applications in computer graphics. Recent studies have demonstrated the advantages of deep neural networks (DNNs) for achieving remarkable performance in motion and music tasks with little effort for feature pre-processing. However, applying DNNs for generating dance to a piece of music is nevertheless challenging, because of 1) DNNs need to generate large sequences while mapping the music input, 2) the DNN needs to constraint the motion beat to the music, and 3) DNNs require a considerable amount of hand-crafted data. In this study, we propose a weakly supervised deep recurrent method for real-time basic dance generation with audio power spectrum as input. The proposed model employs convolutional layers and a multilayered Long Short-Term memory (LSTM) to process the audio input. Then, another deep LSTM layer decodes the target dance sequence. Notably, this end-to-end approach has 1) an auto-conditioned decode configuration that reduces accumulation of feedback error of large dance sequence, 2) uses a contrastive cost function to regulate the mapping between the music and motion beat, and 3) trains with weak labels generated from the motion beat, reducing the amount of hand-crafted data. We evaluate the proposed network based on i) the similarities between generated and the baseline dancer motion with a cross entropy measure for large dance sequences, and ii) accurate timing between the music and motion beat with an F-measure. Experimental results revealed that, after training using a small dataset, the model generates basic dance steps with low cross entropy and maintains an F-measure score similar to that of a baseline dancer.
KW - Contrastive loss
KW - Dance generation
KW - Deep recurrent networks
UR - http://www.scopus.com/inward/record.url?scp=85073227555&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073227555&partnerID=8YFLogxK
U2 - 10.1109/IJCNN.2019.8851872
DO - 10.1109/IJCNN.2019.8851872
M3 - Conference contribution
AN - SCOPUS:85073227555
T3 - Proceedings of the International Joint Conference on Neural Networks
BT - 2019 International Joint Conference on Neural Networks, IJCNN 2019
PB - Institute of Electrical and Electronics Engineers Inc.
ER -