My work is driven by a simple yet ambitious goal: to make weather forecasts as intuitive as looking out your window. Imagine planning a hike and being able to virtually step into the landscape, seeing not just abstract symbols or percentages, but the actual clouds, fog, and sunlight you’ll encounter. That’s the vision behind my research—I’m building a visual twin of the atmosphere and terrain, a realistic 3D environment where anyone can explore forecasted weather in real time.

The challenge? Weather data is enormous, and rendering it realistically—especially in a web browser—demands clever solutions. I’m exploring cutting-edge compression techniques, like tensor decomposition, to shrink these datasets without losing the details that matter, whether for scientific analysis or lifelike visuals. By combining high-fidelity terrain rendering with dynamic atmospheric effects, I aim to create a tool that’s not just powerful, but also accessible on everyday devices. This isn’t just about better technology; it’s about helping people—hikers, farmers, or city planners—see the weather, not just read about it, so they can make smarter decisions with confidence. Curious? Let’s bring the forecast to life!

Disclaimer: This text was generated with the assistance of Le Chat, an AI assistant created by Mistral AI.

I am happy to supervise student projects related to computer graphics and its applications. A selection of available projects in our group can be found here. If none of them sparks your interest, don't hesitate to send me an email, very often I have more interesting projects available or we can discuss your ideas.

• TrailView: An Interactive 3D Hike Planning Tool (2025)
• Informfully - Research Platform for Reproducible User Studies (2024)
• Terrender: A Web-Based Multi-Resolution Terrain Rendering Framework (2022)