Przemyslaw Musialski

My group works at the intersection of geometry, computer graphics, and machine learning. We investigate problems in

geometry processing · shape optimization · digital fabrication · generative AI

My group works where geometry, computer graphics, and machine learning meet. We build generative models for 3D content and design whose output can be edited, built, and trusted, not only looked at. The structure of the domain lives inside the model: the grammar of a facade, the mathematics of a surface that can be manufactured, the physical constraints a real object must satisfy. That is what makes the output controllable for a designer and dependable in the world. It does not compete with scale. A larger model is more capable and less inspectable, and in these domains the need for built-in structure grows with it. The same principle now carries our work into computer vision and the life sciences.