Lightweight Quaternion Transition Generation with Neural Networks

Romi Geleijn; Adrian Radziszewski; Julia Beryl van Straaten; Henrique Galvan Debarba

doi:10.1109/VRW52623.2021.00172

Lightweight Quaternion Transition Generation with Neural Networks

Romi Geleijn, Adrian Radziszewski, Julia Beryl van Straaten, Henrique Galvan Debarba

Research output: Conference Article in Proceeding or Book/Report chapter › Conference abstract in proceedings › Research › peer-review

Abstract

This paper introduces the Quaternion Transition Generator (QTG), a new network architecture tailored to animation transition generation for virtual characters. The QTG is simpler than the current state of the art, making it lightweight and easier to implement. It uses approximately 80% fewer arithmetic operations compared to other transition networks. Additionally, this architecture is capable of generating visually accurate rotation-based animations transitions and results in a lower Mean Absolute Error than transition generation techniques that are commonly used for animation blending.

Original language	English
Title of host publication	2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)
Publisher	IEEE
Publication date	2021
DOIs	https://doi.org/10.1109/VRW52623.2021.00172
Publication status	Published - 2021

Keywords

Quaternion Transition Generator
Animation transition generation
Network architecture
Virtual characters
Rotation-based animations

Access to Document

10.1109/VRW52623.2021.00172

2021-IEEEVR-character_animation_transitionAccepted author manuscript, 1.11 MB

Cite this

@inbook{f69bddb1dc91485fbf68483cf0db3e1b,

title = "Lightweight Quaternion Transition Generation with Neural Networks",

abstract = "This paper introduces the Quaternion Transition Generator (QTG), a new network architecture tailored to animation transition generation for virtual characters. The QTG is simpler than the current state of the art, making it lightweight and easier to implement. It uses approximately 80% fewer arithmetic operations compared to other transition networks. Additionally, this architecture is capable of generating visually accurate rotation-based animations transitions and results in a lower Mean Absolute Error than transition generation techniques that are commonly used for animation blending.",

keywords = "Quaternion Transition Generator, Animation transition generation, Network architecture, Virtual characters, Rotation-based animations",

author = "Romi Geleijn and Adrian Radziszewski and {Beryl van Straaten}, Julia and Debarba, {Henrique Galvan}",

year = "2021",

doi = "10.1109/VRW52623.2021.00172",

language = "English",

booktitle = "2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)",

publisher = "IEEE",

address = "United States",

}

Lightweight Quaternion Transition Generation with Neural Networks. / Geleijn, Romi; Radziszewski, Adrian; Beryl van Straaten, Julia et al.
2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW). IEEE, 2021.

Research output: Conference Article in Proceeding or Book/Report chapter › Conference abstract in proceedings › Research › peer-review

TY - ABST

T1 - Lightweight Quaternion Transition Generation with Neural Networks

AU - Geleijn, Romi

AU - Radziszewski, Adrian

AU - Beryl van Straaten, Julia

AU - Debarba, Henrique Galvan

PY - 2021

Y1 - 2021

N2 - This paper introduces the Quaternion Transition Generator (QTG), a new network architecture tailored to animation transition generation for virtual characters. The QTG is simpler than the current state of the art, making it lightweight and easier to implement. It uses approximately 80% fewer arithmetic operations compared to other transition networks. Additionally, this architecture is capable of generating visually accurate rotation-based animations transitions and results in a lower Mean Absolute Error than transition generation techniques that are commonly used for animation blending.

AB - This paper introduces the Quaternion Transition Generator (QTG), a new network architecture tailored to animation transition generation for virtual characters. The QTG is simpler than the current state of the art, making it lightweight and easier to implement. It uses approximately 80% fewer arithmetic operations compared to other transition networks. Additionally, this architecture is capable of generating visually accurate rotation-based animations transitions and results in a lower Mean Absolute Error than transition generation techniques that are commonly used for animation blending.

KW - Quaternion Transition Generator

KW - Animation transition generation

KW - Network architecture

KW - Virtual characters

KW - Rotation-based animations

U2 - 10.1109/VRW52623.2021.00172

DO - 10.1109/VRW52623.2021.00172

M3 - Conference abstract in proceedings

BT - 2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)

PB - IEEE

ER -

Lightweight Quaternion Transition Generation with Neural Networks

Abstract

Keywords

Access to Document

Fingerprint

Cite this