Thomas Chambon is a Machine Learning research engineer / scientist at Intel Corporation.
His work focuses on machine learning that specifically delves into generative models across various subjects such as denoising diffusion, style transfer, texture synthesis, and super resolution.
Education
| Institution | Degree | Dates |
|---|---|---|
| INSA de Lyon | Master of Computer Science (French title: Diplome d’ingénieur INSA de Lyon, avec félicitations du jury) | Sept 2006 - June 2011 |
Experience
| Company | Role | Dates |
|---|---|---|
| Intel Corporation | Machine Learning Research Engineer / Scientist | 2023/02 - present |
| Unity Technologies | Senior Machine Learning Engineer | 2020/01 - 2023/02 |
| Freelance | Machine Learning Engineer | 2018/09 - 2020/01 |
| EDF | Lead Data architect | 2015/08 - 2018/09 |
| Wavestone | Senior Data consultant | 2011/04 - 2015/08 |
Publications
Iterative 𝛼-(de)Blending: a Minimalist Deterministic Diffusion Model
Eric Heitz, Laurent Belcour and Thomas Chambon
Published in ACM SIGGRAPH 2023
The Energy Distance as a Replacement for the Reconstruction Loss in Conditional GANs
Eric Heitz and Thomas Chambon
Published in JCGT 2023
Passing Multi-Channel Material Textures to a 3-Channel Loss
Thomas Chambon, Eric Heitz and Laurent Belcour
Published in ACM SIGGRAPH Talk 2021
A Sliced Wasserstein Loss for Neural Texture Synthesis
Eric Heitz, Kenneth Vanhoey, Thomas Chambon, Laurent Belcour
Published in CVPR 2021
