Bonjour, Hello!

I am a Ph.D. candidate at the McGill Graphics Lab (MGL), working under Derek Nowrouzezahrai. I am currently interning at NVIDIA AI as a Research Scientist in Deep Learning, where I have joined Sanja Fidler's group.

My main research interests lie at the intersection of physically-based rendering and machine learning. More precisely, I am interested in applying deep learning techniques to improve current rendering algorithms (e.g., denoising, path guiding, sampling). I also have an interest for differentiable rendering, 3D representation learning and Markov chain Monte Carlo (MCMC) methods for light transport simulations.

If you'd like to chat, feel free to send me an email.

Latest News (See more )

2021.03.03
Neural LOD has been accepted for oral presentation at CVPR 2021.
2021.01.26
Our new paper from my research internship at NVIDIA is now online!
2020.12.11
I am helping to organize GRAPHQUON (previously MOTOGRAPH), which will be virtual this year.

Recent Publications (See more )

Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes

T. Takikawa, J. Litalien, K. Yin, K. Kreis, C. Loop, D. Nowrouzezahrai, A. Jacobson, M. McGuire, and S. Fidler
Computer Vision and Pattern Recognition (CVPR), January 2021

Selected Projects (See more )

Online Test Suite for Rendering Research

A web comparison tool for rendering research written in Python+JS, based on the online test suite with interactive viewer by Disney Research. Useful for assembling supplementary material (comparison metrics, convergence plots, etc.)

Noise2Noise PyTorch Implementation

Unofficial PyTorch implementation of Noise2Noise: Learning Image Restoration without Clean Data by Lehtinen et al., 2018