Cyprien Plateau–Holleville

Cyprien
Plateau--Holleville

Postdoctoral Researcher in Computer Graphics at Inria

I'm a French PhD in Computer Graphics currently working as a postdoctoral researcher for Fast Fluid Simulation based on Optimal Transport at Inria and under the supervision of Bruno Lévy. Before that, I completed a PhD in Computer Graphics at the University of Limoges under the supervision of Stéphane Mérillou and Maxime Maria during which I worked for Efficient Computation of Geometry for Large Protein Visualization and Illustration.

Motivations
My research focuses on efficient geometric representations and parallel computational methods. In particular, I'm interested in large-scale applications in rendering, scientific visualization and physics simulation.

Contact: cyprien [dot] plateau-holleville [at] inria [dot] fr

Work In Progress

Fluid Simulation based on Optimal Transport

Start of a fluid simulation relying on Partial Optimal Transport. This model guarantees that the fluid keeps the prescribed volume while offering an interesting explicit characterization of the geometry. The underlying representation (central part) is analytically intersected and rendered using a pathtracer.

Academic Publications

VTX: Real-time high-performance molecular structure and dynamics visualization software

Maxime Maria; Simon Guionnière; Nicolas Dacquay; Cyprien Plateau-Holleville; Valentin Guillaume; Vincent Larroque; Jean Lardé; Yassine Naimi; Jean-Philip Piquemal; Guillaume Levieux; Nathalie Lagarde; Stéphane Mérillou; Matthieu Montes

2025 Bioinformatics

PDF (958KB) Project page

In Search of Empty Spheres: 3D Apollonius Diagrams on GPU

Cyprien Plateau-Holleville; Benjamin Stamm; Vincent Nivoliers; Maxime Maria; Stéphane Mérillou

2025 SIGGRAPH (ACM Transactions on Graphics)

PDF (51MB) PDF compressed (14.5MB) Project page

2025

PhD thesis

Efficient computation of geometry for structural analysis of large molecular systems

TLDR: Efficient computation of Molecular Surface; Apollonius diagrams characterization and parametrization; GPU Apollonius diagrams computation.

PDF (270MB) PDF compressed (30.9MB) Video

Dissertation Award (GDR-IGRV)

Construction de diagrammes d’Apollonius 3D sur GPU

Cyprien Plateau-Holleville; Benjamin Stamm; Vincent Nivoliers; Maxime Maria; Stéphane Mérillou

Journées Française d’Informatique Graphique 2024 (jFIG)

PDF (31.65MB)

Best Paper Award

Efficient GPU computation of large protein Solvent-Excluded Surface

Cyprien Plateau-Holleville; Maxime Maria; Stéphane Mérillou; Matthieu Montes

2024 IEEE Transactions on Visualization and Computer Graphics

PDF (36.6MB) PDF compressed (1MB) Supplemental Project page

2024

UDock2: Interactive real-time multi-body protein-protein docking software

Cyprien Plateau-Holleville; Simon Guionnière; Benjamin Boyer; Brian Jiménez-Garcia; Guillaume Levieux; Stéphane Mérillou; Maxime Maria; Matthieu Montes

2023 Bioinformatics

PDF Project page Source code

2023

Class-aware data augmentation by GAN specialisation to improve endoscopic images classification

Cyprien Plateau-Holleville; Yannick Benezeth

2022 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI)

PDF Source code

2022

French vital records data gathering and analysis through image processing and machine learning algorithms

Cyprien Plateau-Holleville^*; Enzo Bonnot^*; Franck Gechter; Laurent Heyberger

2021 Journal of Data Mining & Digital Humanities

^*Both authors contributed the same to this work.

PDF Source code

Side Projects

I like to have side projects to test things that can support my research. Some of the following projects have laid the fundations for the rendering engines used in my publications.

Shadertoy

I'm using Shadertoy a lot to implement papers and test methods. I also participated to the Shadertoy contest of the French Computer Graphics association (AFIG) between 2021 and 2023. My last entry has been awarded with the first prize while both 2021 and 2022 entries won the second prizes.

Profile Toriversaire (2023) Cannelés (2022) JellyFish (2021)

Vazteran

Vulkan library used as a sandbox for my graphics project. The image is rendered in real time with 1M particles with a Macbook Pro M1 (see the corresponding example app/particles).

Source code

Launcher of Particle

Path-Tracing engine mainly planned to test high quality rendering algorithms based on Vazteran. It can currently render meshes with a customizable material based on Disney BSDF with environment map MIS. The current goal is to clean up the API to allow more configuration.

Source code

Mirage

An image processing library for learning purpose in C++17. This project supported my studies in computer vision and I don't currently plan to add more features.

Source code