Justin Johnson

Justin Johnson is an Assistant Professor at the University of Michigan's College of Engineering, Department of Electrical Engineering and Computer Science, and a Research Scientist at Facebook AI Research (FAIR). His work bridges computer vision, machine learning, and deep learning, focusing on visual reasoning, vision-language tasks, image generation, and 3D reasoning using neural networks.

• PhD, Stanford University (advised by Fei-Fei Li)

His research interests span visual reasoning, vision and language, 3D vision, and image generation, with a focus on innovative applications of deep neural networks. Recent publications highlight work on 3D consistency, self-supervised learning, and multimodal integration of vision and text.

Notable contributions include PyTorch3D for 3D data processing, and foundational work in visual question answering, neural style transfer, and scene graph-based image generation. Publications span top conferences like ICCV, CVPR, and NeurIPS.

He teaches courses including EECS 498/598: Deep Learning for Computer Vision and EECS 442: Computer Vision at University of Michigan, with prior involvement in Stanford's CS 231N in co-teaching roles.

Software projects include open-source frameworks like fast-neural-style for real-time artistic style transfer, and PyTorch3D for efficient 3D deep learning. These tools demonstrate his commitment to practical implementations and community-driven research.