Neuranimation: Reactive Character Animations with Deep Neural Networks

Sebastian Silva; Sergio Sugahara; Willy Ugarte

doi:10.5220/0010896500003124

Neuranimation: Reactive Character Animations with Deep Neural Networks

Sebastian Silva
, Sergio Sugahara
, Willy Ugarte

Faculty of Engineering

Universidad Peruana de Ciencias Aplicadas

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

The increasing need for more realistic animations has resulted in the implementation of various systems that try to overcome this issue by controlling the character at a base level based on complex techniques. In our work we are using a Phase Functioned Neural Network for generating the next pose of the character in real time while making a comparison with a modified version of the model. The current basic model lacks the ability of producing reactive animations with objects of their surrounding but only reacts to the terrain the character is standing on. Therefore, adding a layer of Rigs with Inverse Kinematics and Blending Trees will allow us to switch between actions depending on the object and adjust the character to fit properly. Our results showed that our proposal improves significantly previous results and that inverse kinematics is essential for this improvement.

Original language	English
Title of host publication	VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1
Editors	A. Augusto de Sousa, Kurt Debattista, Kadi Bouatouch
Publisher	Science and Technology Publications, Lda
Pages	252-259
Number of pages	8
ISBN (Print)	9789897585555
DOIs	https://doi.org/10.5220/0010896500003124
State	Published - 2022
Event	17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022 - Virtual, Online Duration: 6 Feb 2022 → 8 Feb 2022

Publication series

Name	Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Volume	1
ISSN (Print)	2184-5921
ISSN (Electronic)	2184-4321

Conference

Conference	17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022
City	Virtual, Online
Period	6/02/22 → 8/02/22

Keywords

Character Animation
Character Control
Deep Learning
Human Motion
Locomotion
Neural Networks

Access to Document

10.5220/0010896500003124

Cite this

Silva, S., Sugahara, S., & Ugarte, W. (2022). Neuranimation: Reactive Character Animations with Deep Neural Networks. In A. A. de Sousa, K. Debattista, & K. Bouatouch (Eds.), VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1 (pp. 252-259). (Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications; Vol. 1). Science and Technology Publications, Lda. https://doi.org/10.5220/0010896500003124

Silva, Sebastian ; Sugahara, Sergio ; Ugarte, Willy. / Neuranimation : Reactive Character Animations with Deep Neural Networks. VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1. editor / A. Augusto de Sousa ; Kurt Debattista ; Kadi Bouatouch. Science and Technology Publications, Lda, 2022. pp. 252-259 (Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications).

@inproceedings{7f9768df60094ede8878ceff357cdf43,

title = "Neuranimation: Reactive Character Animations with Deep Neural Networks",

abstract = "The increasing need for more realistic animations has resulted in the implementation of various systems that try to overcome this issue by controlling the character at a base level based on complex techniques. In our work we are using a Phase Functioned Neural Network for generating the next pose of the character in real time while making a comparison with a modified version of the model. The current basic model lacks the ability of producing reactive animations with objects of their surrounding but only reacts to the terrain the character is standing on. Therefore, adding a layer of Rigs with Inverse Kinematics and Blending Trees will allow us to switch between actions depending on the object and adjust the character to fit properly. Our results showed that our proposal improves significantly previous results and that inverse kinematics is essential for this improvement.",

keywords = "Character Animation, Character Control, Deep Learning, Human Motion, Locomotion, Neural Networks",

author = "Sebastian Silva and Sergio Sugahara and Willy Ugarte",

note = "Publisher Copyright: {\textcopyright} 2022 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved; 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022 ; Conference date: 06-02-2022 Through 08-02-2022",

year = "2022",

doi = "10.5220/0010896500003124",

language = "Ingl{\'e}s",

isbn = "9789897585555",

series = "Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications",

publisher = "Science and Technology Publications, Lda",

pages = "252--259",

editor = "\{de Sousa\}, \{A. Augusto\} and Kurt Debattista and Kadi Bouatouch",

booktitle = "VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1",

}

Silva, S, Sugahara, S & Ugarte, W 2022, Neuranimation: Reactive Character Animations with Deep Neural Networks. in AA de Sousa, K Debattista & K Bouatouch (eds), VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1. Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, vol. 1, Science and Technology Publications, Lda, pp. 252-259, 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022, Virtual, Online, 6/02/22. https://doi.org/10.5220/0010896500003124

Neuranimation: Reactive Character Animations with Deep Neural Networks. / Silva, Sebastian; Sugahara, Sergio; Ugarte, Willy.
VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1. ed. / A. Augusto de Sousa; Kurt Debattista; Kadi Bouatouch. Science and Technology Publications, Lda, 2022. p. 252-259 (Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications; Vol. 1).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Neuranimation

T2 - 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2022

AU - Silva, Sebastian

AU - Sugahara, Sergio

AU - Ugarte, Willy

PY - 2022

Y1 - 2022

N2 - The increasing need for more realistic animations has resulted in the implementation of various systems that try to overcome this issue by controlling the character at a base level based on complex techniques. In our work we are using a Phase Functioned Neural Network for generating the next pose of the character in real time while making a comparison with a modified version of the model. The current basic model lacks the ability of producing reactive animations with objects of their surrounding but only reacts to the terrain the character is standing on. Therefore, adding a layer of Rigs with Inverse Kinematics and Blending Trees will allow us to switch between actions depending on the object and adjust the character to fit properly. Our results showed that our proposal improves significantly previous results and that inverse kinematics is essential for this improvement.

AB - The increasing need for more realistic animations has resulted in the implementation of various systems that try to overcome this issue by controlling the character at a base level based on complex techniques. In our work we are using a Phase Functioned Neural Network for generating the next pose of the character in real time while making a comparison with a modified version of the model. The current basic model lacks the ability of producing reactive animations with objects of their surrounding but only reacts to the terrain the character is standing on. Therefore, adding a layer of Rigs with Inverse Kinematics and Blending Trees will allow us to switch between actions depending on the object and adjust the character to fit properly. Our results showed that our proposal improves significantly previous results and that inverse kinematics is essential for this improvement.

KW - Character Animation

KW - Character Control

KW - Deep Learning

KW - Human Motion

KW - Locomotion

KW - Neural Networks

UR - https://www.scopus.com/pages/publications/85180770370

U2 - 10.5220/0010896500003124

DO - 10.5220/0010896500003124

M3 - Contribución a la conferencia

AN - SCOPUS:85180770370

SN - 9789897585555

T3 - Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

SP - 252

EP - 259

BT - VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1

A2 - de Sousa, A. Augusto

A2 - Debattista, Kurt

A2 - Bouatouch, Kadi

PB - Science and Technology Publications, Lda

Y2 - 6 February 2022 through 8 February 2022

ER -

Silva S, Sugahara S, Ugarte W. Neuranimation: Reactive Character Animations with Deep Neural Networks. In de Sousa AA, Debattista K, Bouatouch K, editors, VISIGRAPP 2022 - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 1. Science and Technology Publications, Lda. 2022. p. 252-259. (Proceedings of the International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications). doi: 10.5220/0010896500003124

Neuranimation: Reactive Character Animations with Deep Neural Networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this