Deep Learning for Fluid Simulation and Animation: Fundamentals, Modeling, and Case Studies (SpringerBriefs in Mathematics) - Softcover

Giraldi, Gilson Antonio; Almeida, Liliane Rodrigues De; Apolinário Jr., Antonio Lopes; Silva, Leandro Tavares Da

Softcover

ISBN 10: 3031423321 ISBN 13: 9783031423321

Publisher: Springer, 2023

Synopsis

This book is an introduction to the use of machine learning and data-driven approaches in fluid simulation and animation, as an alternative to traditional modeling techniques based on partial differential equations and numerical methods – and at a lower computational cost.

This work starts with a brief review of computability theory, aimed to convince the reader – more specifically, researchers of more traditional areas of mathematical modeling – about the power of neural computing in fluid animations. In these initial chapters, fluid modeling through Navier-Stokes equations and numerical methods are also discussed.

The following chapters explore the advantages of the neural networks approach and show the building blocks of neural networks for fluid simulation. They cover aspects related to training data, data augmentation, and testing. 

The volume completes with two case studies, one involving Lagrangian simulation of fluids using convolutional neural networks and the other using Generative Adversarial Networks (GANs) approaches.

About this title

About the Author

Gilson Antonio Giraldi is a Researcher at the National Laboratory for Scientific Computing (LNCC), Brazil, where he is responsible for academic research projects in image analysis, statistical and machine learning, scientific visualization, and physically-based animation. He holds a PhD in Computer Graphics (2000) from the Federal University of Rio de Janeiro, Brazil, and has a degree in Mathematics (1986) from the Pontifical Catholic University of Campinas, Brazil.

Antonio Lopes Apolinário Junior is an Associate Professor at the Federal University of Bahia (UFBA), Brazil. He holds a PhD in Systems and Computer Engineering (2004) from the Federal University of Rio de Janeiro, Brazil. His research interests lie in computer graphics, 3D modeling, augmented reality, virtual reality, and physically-based rendering and animation.

Leandro Tavares da Silva is a Professor at the Federal Center for Technological Education “Celso Suckow da Fonseca” (CEFET-RJ), Brazil. He holds a PhD in Computational Modeling (2016) from the National Laboratory for Scientific Computing (LNCC), Brazil. He currently does research on fluid simulation and animation, and deep learning. 

Liliane Rodrigues de Almeida is a Fellow Researcher at the National Laboratory for Scientific Computing (LNCC). She holds a Master’s degree in Computer Science (2017) from the Federal University of Juiz de Fora (UFJF), Brazil, and has a degree in Computer Science from the same university. Her fields of research are physical simulation and computational geometry.

From the Back Cover

This book is an introduction to the use of machine learning and data-driven approaches in fluid simulation and animation, as an alternative to traditional modeling techniques based on partial differential equations and numerical methods – and at a lower computational cost.

This work starts with a brief review of computability theory, aimed to convince the reader – more specifically, researchers of more traditional areas of mathematical modeling – about the power of neural computing in fluid animations. In these initial chapters, fluid modeling through Navier-Stokes equations and numerical methods are also discussed.

The following chapters explore the advantages of the neural networks approach and show the building blocks of neural networks for fluid simulation. They cover aspects related to training data, data augmentation, and testing. 

The volume completes with two case studies, one involving Lagrangian simulation of fluids using convolutional neural networks and the other using Generative Adversarial Networks (GANs) approaches.

Edition details

• Publisher: Springer
• Publication date: 2023
• ISBN 10: 3031423321
• ISBN 13: 9783031423321
• Binding: Paperback
• Language: English
• Number of pages: 180