I am a recent MSc Computer Science graduate from TU Delft. My research interests lie in the intersection of computer graphics, deep learning and computer vision. I am particularly interested in the application of deep learning for computer graphics tasks (like this or this) and in the combination of classical graphics ideas and data-driven computer vision for capturing environments (like this and this).

For my Master’s thesis, I worked on joint reconstruction of object appearance, geometry and physics parameters with the mentorship of Prof. Petr Kellnhofer. In the past, I have worked on physically based rendering with Prof. Elmar Eisemann, intercontinental robotic teleoperation with Prof. VP and static analysis for software engineering with Prof. Andy Zaidman.

I enjoy scuba diving and I am certified PADI Rescue Diver. I also love hiking but there are no mountains in the Netherlands.

Selected (Research) Projects

Here are some of the projects I have proudly done or participated in.

Ambient Light Caching via Approximate Photon Mapping

Researched and developed a photon mapping-based approach to create ambient light caches to improve the efficiency of physically based rendering. See here for the report.

The Unnamed Game Engine

OpenGL 3D game engine built with the ENTT entity component system with support for shadow mapping, physically based materials, SDF glyph rendering and Bézier curve animations. Check it out!

Visualizing the Gaussian Splatting Optimization

A 3D visualizer that demonstrates how the 3D Gaussian Splatting optimization algorithm adapts the Gaussian Kernels to match the underlying scene’s appearance and geometry. Click here to see it in action and here for the code.

Visualizing Practical Path Guiding

An interactive visualization of Practical Path Guiding by Müller, Gross and Novák. We reimplement their technique in a custom 2D path tracer, reducing the radiance field and the proposed adaptive spatio-directional tree to three dimensions (instead of five). Since the tree and radiance field are 3D we can then visualize how the first adapts to fit the second.

Blender Geometry Processing

A Blender 3D plugin that implements various geometry processing algorithms. Specifically I worked on implementing and evaluating different variants of the Iterative Closest Point algorithm and numerous mesh editing brushes based on Laplacian and gradient transformatoins. See here for a demo and here for the code.

TestKnight

Published in ICSE 2022. An automated tool that assists in software testing engineering. Developed as a plugin for the Intellij Idea, available in the JetBrains Marketplace. Code available here.