currently

I am a phd student at INSAIT, supervised by Danda Pani Paudel and Luc van Gool. I work at the intersection of machine learning, computer vision and robotics. I am interested in building general purpose agents that autonomously solve tasks in the real world. Currently, I am exploring how to connect spatial representation (or memory) of the world with control policies like VLAs or VAMs to reason and interact with the world. If you are interested in these topics, I encourage you to reach out via email.


previously

I completed both my Bachelor’s and Master’s at TUM. I started out in machine learning, working on robustness topics, then switched to 3D vision, where I worked on shape generation, 3D reconstruction, and inverse rendering. Along the way, I was fortunate to spend time at KAIST in South Korea, ETH Zürich in Switzerland, and the University of Cambridge in the UK, and to work closely with many great people, including Matthias Nießner, Abhi Meka, and Ayush Tewari. Besides my studies, I was active at TUM.ai, where I helped organise events to grow the AI ecosystem in Munich.

contact

You can reach me via email: daniel.korth@insait.ai.