Peter is a Research Scientist at NVIDIA’s real-tim graphics group, joined in April 2026. His research focuses on generative inverse and forward rendering, leveraging large-scale image priors to solve fundamental challendes in material estimation and tractable light transport. Peter did his PhD at the Technical University of Munich and he was interning at Adobe Research. Before pivoting to computer graphics, he worked on general computer vision projects, such as active learning and finished his studies in the field of engineering, focusing on neural control for robotics and autonomous vehicles. Peter is excited about building systems that represent the world and bridging gap between simulation and reality.

A complete list of prior publications can be found here.