معرفی
Marcus A. Brubaker is an Associate Professor of Computer Science at York University, founder and director of the Computational Vision and Imaging Lab (CVIL), Faculty Affiliate at the Vector Institute, and Adjunct Professor at the University of Toronto Department of Computer Science. He holds additional roles as a Visiting Professor at Samsung AI Center - Toronto, co-founder and advisor of Structura Biotechnology (creators of cryoSPARC), and Academic Advisor at Borealis AI.
His research focuses on building rich models capturing fundamental relationships between the world and observations, with recent emphasis on generative models (particularly normalizing flows) and Cryo-EM structure estimation of biological molecules. His broader interests span vehicle localization, physically realistic human motion modeling, probabilistic programming languages, Bayesian methods, MCMC, and forensic ballistics.
His publication trends reveal strong specialization in computer vision and generative modeling, with recent ICCV 2023 papers advancing neural radiance field inpainting and photometrically consistent view synthesis using diffusion models. These works demonstrate interdisciplinary integration of computer graphics, deep learning, and novel view generation techniques.
Brubaker has served as Associate Editor for IET Computer Vision and Area Chair/Senior Area Chair for major conferences including CVPR, ECCV, NeurIPS, and AAAI. Previously, he was Research Director at Borealis AI (2018-2020), Research Associate at Cadre Research Labs (creators of TopMatch-GS), and an original contributor to the Stan probabilistic programming framework.
He directs the Computational Vision and Imaging Lab (CVIL) at York University alongside colleagues Kosta Derpanis and Michael Brown, focusing on cutting-edge computer vision research with applications in biotechnology and AI. As of August 2024, he has commenced a leave of absence from York University to join Google DeepMind as a Research Scientist in Toronto.
