How to train your (neural) dragon

被引:0
作者
Schirmer, Luiz [1 ]
Novello, Tiago [2 ]
da Silva, Vinicius [3 ]
Schardong, Guilherme [4 ]
Lopes, Helio [3 ]
Velho, Luiz [2 ]
机构
[1] Univ Vale Rio dos Sinos, Sao Leopoldo, Brazil
[2] IMPA, Rio De Janeiro, Brazil
[3] Pontificia Univ Catolica Rio de Janeiro, Rio de Janeiro, Brazil
[4] Univ Coimbra, Coimbra, Portugal
来源
2023 36TH CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES, SIBGRAPI 2023 | 2023年
关键词
neural fields; differential geometry;
D O I
10.1109/SIBGRAPI59091.2023.10347177
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Neural fields have emerged as a promising framework for representing different types of signals. This tutorial focus on the existing literature and shares practical insights derived from hands-on experimentation with neural fields, specifically in approximating implicit functions of surfaces. Our emphasis lies in strategies leveraging differential geometry concepts to enhance training outcomes and showcase applications within this domain.
引用
收藏
页码:264 / 269
页数:6
相关论文
共 16 条
[11]  
Parascandolo G., 2017, Taming the waves: sine as activation function in deep neural networks
[12]   Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations [J].
Raissi, M. ;
Perdikaris, P. ;
Karniadakis, G. E. .
JOURNAL OF COMPUTATIONAL PHYSICS, 2019, 378 :686-707
[13]   Neural Networks for Implicit Representations of 3D Scenes [J].
Schirmer, Luiz ;
Schardong, Guilherme ;
da Silva, Vinicius ;
Lopes, Helio ;
Novello, Tiago ;
Yukimura, Daniel ;
Magalhaes, Thales ;
Paz, Hallison ;
Velho, Luiz .
2021 34TH SIBGRAPI CONFERENCE ON GRAPHICS, PATTERNS AND IMAGES (SIBGRAPI 2021), 2021, :17-24
[14]  
Sitzmann Vincent, 2020, Advances in Neural Information Processing Systems, V33
[15]  
Takikawa T., 2021, arXiv
[16]  
TAUBIN G, 1995, FIFTH INTERNATIONAL CONFERENCE ON COMPUTER VISION, PROCEEDINGS, P902, DOI 10.1109/ICCV.1995.466840