Kiwon Um

Kiwon Um, Ph.D.

Assistant Professor (Maître de Conférences in French)

Computer Graphics Group, IMAGES Team, IDS Department, LTCI

My research interests include physics-based simulations and data-driven approaches with deep learning as well as perceptual evaluation of simulations. I have been investigating more effective ways to simulate natural phenomena via utilization of refined data and to acquire reliable data for machine learning. Moreover, I have been studying human visual perception for not only computer graphics but also engineering applications while aiming for new evaluation approaches and better understanding of a variety of numerical methods.

I am an assistant professor in the Computer Graphics group at Telecom Paris, France. Before joining Telecom Paris in October 2019, I worked as a postdoctoral researcher at the Technical University of Munich with Nils Thuerey and Korea University from which I earned my Ph.D. in computer science and engineering under JungHyun Han’s supervision.

Contact

E-mail: kiwon.um at telecom-paris dot fr
Phone: +33 (0)1 75 31 96 29 (Office)
Office: 5.B20

Supervision

Francesco Fossella, PhD student (since 10/2023)

co-supervised with Mathieu Desbrun (INRIA & École Polytechnique) and Luca Biferale (University of Rome, Tor Vergata)

Elisa Bellantoni, PhD student (since 10/2023)

co-supervised with Mathieu Desbrun (INRIA & École Polytechnique), Mauro Sbragaglia (University of Rome, Tor Vergata), and Nikos Savva (The Cyprus Institute)

André Freitas, PhD student (since 11/2023)

co-supervised with Mathieu Desbrun (INRIA & École Polytechnique) and Luca Biferale (University of Rome, Tor Vergata)

Alumni

Jiong Chen, Postdoc (09/2021-08/2022), co-supervised with Mathieu Desbrun (Inria & Ecole Polytechnique)
Chloé Paliard, PhD (10/2020-06/2024), co-supervised with Nils Thuerey (TUM) and Marco Cagnazzo (University of Padua)

Openings

There is no openning at this moment.

Teaching

Telecom Paris

Sorbonne University

IG3D (M1) - 3D Computer Graphics

Codes

Solver-in-the-Loop—PDE correction model using a differentiable solver interacting within the training process
MLFLIP—Simple implementation of a liquid splash model with neural networks
Mantaflow—Open-source framework for fluid simulation in computer graphics

Publications

Papers

Solver-in-the-loop Approach to Turbulence Closure

André Freitas, Kiwon Um, Mathieu Desbrun, Michele Buzzicotti, and Luca Biferale

arXiv preprint, 2024.

Momentum-preserving Inversion Alleviation for Elastic Material Simulation

Heejo Jeong, Seung-wook Kim, JaeHyun Lee, Kiwon Um, Min Hyung Kee, and JungHyun Han

Computer Animation and Virtual Worlds (the 37th International Conference on Computer Animation and Social Agents (CASA), June 5-7, 2024, Wuhan, China). Vol. 35, No. 3, pp. e2249.

Exploring Physical Latent Spaces for High-Resolution Flow Restoration

Chloé Paliard, Nils Thuerey, and Kiwon Um

Proceedings of the 28th International Symposium on Vision, Modeling, and Visualization (VMV), September 27-29, 2023, Braunschweig, Germany, pp. 199-207.

Inversion Alleviation for Stable Elastic Body Simulation

JaeHyun Lee, Seung-wook Kim, Kiwon Um, Min Hyung Kee, and JungHyun Han

Computer Animation and Virtual Worlds (the 36th International Conference on Computer Animation and Social Agents (CASA), May 29-31, 2023, Limassol, Cyprus). Vol. 34, No. 3-4, pp. e2183
Video (YouTube)

An Optimization-based SPH Solver for Simulation of Hyperelastic Solids

Min Hyung Kee, Kiwon Um, Hyunmo Kang, and JungHyun Han

Computer Graphics Forum (the 48th Eurographics conference, May 8-12, 2023, Saarbrücken, Germany). Vol. 42, No. 2, pp. 225-233.
Video (YouTube)

Constrained Projective Dynamics: Real-Time Simulation of Deformable Objects with Energy-Momentum Conservation

Min Hyung Kee, Kiwon Um, WooSeok Jeong, and JungHyun Han

ACM Transactions on Graphics (the 48th SIGGRAPH Conference, August 9-13, 2021). Vol. 40, No. 4, Article 160.
Teaser (YouTube), Video (YouTube)

Spot the Difference: Accuracy of Numerical Simulations via the Human Visual System

Kiwon Um, Xiangyu Hu, Bing Wang, and Nils Thuerey

ACM Transactions on Applied Perception (May 2021). Vol. 18, No. 2, Article 6.

Solver-in-the-Loop: Learning from Differentiable Physics to Interact with Iterative PDE-Solvers

Kiwon Um, Robert Brand, Yun (Raymond) Fei, Philipp Holl, and Nils Thuerey

Proceedings of Advances in Neural Information Processing Systems (NeurIPS), December 6-12, 2020, Vancouver, Canada. Vol. 33, pp. 6111-6122.

Georg Kohl, Kiwon Um, and Nils Thuerey

Proceedings of the 37th International Conference on Machine Learning (ICML), July 12-18, 2020.

ScalarFlow: A Large-Scale Volumetric Data Set of Real-world Scalar Transport Flows for Computer Animation and Machine Learning

Marie-Lena Eckert, Kiwon Um, and Nils Thuerey

ACM Transactions on Graphics (the 12th SIGGRAPH Asia Conference, November 17-20, 2019, Brisbane, Australia). Vol. 38, Issue 6, Article 239.

Liquid Splash Modeling with Neural Networks

Kiwon Um, Xiangyu Hu, and Nils Thuerey

Computer Graphics Forum (the 17th annual Symposium on Computer Animation, July 11-13, 2018, Paris, France), Vol. 37, No. 8, pp. 171-182.

Perceptual Evaluation of Liquid Simulation Methods

Kiwon Um, Xiangyu Hu, and Nils Thuerey

ACM Transactions on Graphics (the 44th SIGGRAPH Conference, July 30-August 3, 2017, Los Angeles, USA), Vol. 36, Issue 4, Article 143.

Correspondence-based Fluid Control with a Hybrid Particle-Grid Simulation Method

Kiwon Um, Seungho Baek, and JungHyun Han

Proceedings of the 30th International Conference on Computer Animation and Social Agents, May 22-24, 2017, Seoul, Korea. (short paper)

Analysis of Free Surface Simulation using Breaking-dam Benchmark

Kiwon Um, Xiangyu Hu, and Nils Thuerey

Proceedings of the 11th International SPHERIC Workshop, June 13–16, 2016, Munich, Germany, pp. 410-415.

Muddy Water Animation with Different Details

Seungho Baek, Kiwon Um, and JungHyun Han

Computer Animation and Virtual Worlds (the 28th International Conference on Computer Animation and Social Agents, May 11-13, 2015, Singapore), Vol. 26, No. 3-4, pp. 347-355.

Advanced Hybrid Particle-Grid Method with Sub-Grid Particle Correction

Kiwon Um, Seungho Baek, and JungHyun Han

Computer Graphics Forum (the 22nd Pacific Graphics, October 8-10, 2014, Seoul, Korea), Vol. 33, No. 7, pp. 209-218.

Computer-generated Iron Filing Art

Wonbae Yoon, Namil Lee, Kiwon Um, and JungHyun Han

The Visual Computer (the 31st Computer Graphics International, June 10-13, 2014, Sydney, Australia), Vol. 30, No. 6-8, pp. 889-895.

Porous Deformable Shell Simulation with Surface Water Flow and Saturation

Kiwon Um, Tae-Yong Kim, Youngdon Kwon, and JungHyun Han

Computer Animation and Virtual Worlds (the 26th International Conference on Computer Animation and Social Agents, May 16-18, 2013, Istanbul, Turkey), Vol. 24, No. 3-4, pp. 247-254.

Leaking Fluids

Kiwon Um and JungHyun Han

Proceedings of International Symposium on Visual Computing, December 1-3, 2008, Las Vegas, Nevada, USA, pp. 1135-1143.

Book

Physics-based Deep Learning

Nils Thuerey, Philipp Holl, Maximilian Mueller, Patrick Schnell, Felix Trost, and Kiwon Um

Online version, Nils' introductory video (YouTube)

Theses

Kiwon Um, "Fluid Simulation in Hybrid Frameworks and Porous Shell Simulation with Water," Ph.D. Thesis, Korea University, February 2014. (English)
Kiwon Um, "Leaking Fluids," M.Sc. Thesis, Korea University, February 2008. (Korean)