Currently finishing my studies in computer graphics at ESIR school, I'm doing an internship as a Graphics research intern at Unity Technologies. I'm particularly interested in light transport simulation and image synthesis tools but I also worked on image processing and computer vision projects. The purpose of this page is to list and present some of the projects I have done.
The main projects I have worked on concern light transport simulation and related mathematical optimizations. In particular, I worked on the Monte Carlo error distribution as blue noise in screen space [Heitz and Belcour 2019]. The objective of this tool is to manipulate the random sequences used by the rendering engine in order to get an image whose error follows a high frequency noise. The high frequency noise has interesting visual properties as well as denoising perspectives.
I have also worked on Optimal Multiple Importance Sampling [Kondapaneni et al. 2019] and its implementation in bidirectional path tracing (BDPT). Multiple importance sampling (MIS) is a crucial aspect of BDPT and allows to greatly improve the quality of the results obtained. The optimal version of this tool is based on the result of complex linear systems requiring a fine understanding of light transport theory. A comparative example of the standard MIS method and the optimal one is shown above.
Finally, in addition to my projects in image synthesis, I worked on some color and style transfer tools. This project was for me an opening on a more artistic aspect of computer graphics. You can see an example of the result of this project.
Feel free to contact me if you have some questions.