Summary
Hubert Kompanowski is a PhD researcher and Research Intern based in Ireland with eight years of industry and research experience at the intersection of generative AI, computer vision, and 3D graphics. He develops controllable text-to-3D and multiview diffusion methods—authoring the Stylized Score Distillation technique featured in Dream-in-Style (3DV 2025) and delivering measurable gains in human-evaluation metrics. His work spans both applied engineering (custom Blender pipelines and production-ready multiview datasets) and foundational model adaptation across projects like DreamFusion and Era3D. Prior roles in vision engineering and software development (including IBM and several vision startups) give him a pragmatic lens for moving research into prototypes and products. Unusually for a PhD student, he combines deep generative-model innovation with hands-on tooling for PBR material control, making his outputs immediately useful for 3D design workflows.
8 years of coding experience
4 years of employment as a software developer
Erasmus+ programme, Computer Science, Erasmus+ programme, Computer Science at University of Twente
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Trinity College Dublin
Master's degree, Artificial Intelligence, Master's degree, Artificial Intelligence at AGH University of Krakow
English, Polish