Jan Frahm

Jan-Michael Frahm is a Research Professor in the Department of Computer Science at the University of North Carolina at Chapel Hill. He earned his Ph.D. in 2005 from Christian-Albrechts-University Kiel, Germany. After completing his doctorate, Frahm joined the University of North Carolina at Chapel Hill as a postdoctoral researcher in the computer vision group under Marc Pollefeys. In 2007, he was appointed Research Assistant Professor, later advancing to his current position as Research Professor. Frahm co-heads the 3D Computer Vision research group and serves as a founding faculty member of the Consortium for Vision and Virtual Reality (CV²R). His career at UNC has focused on advancing computer vision technologies, with contributions to projects funded by the National Science Foundation.

Frahm's research interests encompass structure from motion, camera self-calibration, camera sensor systems, multi-camera systems, multi-view stereo, robust estimation, fast tracking of salient features in images and video, computer vision, active vision for model improvement, and markerless augmented reality. He has made seminal contributions to large-scale 3D reconstruction from unordered internet photo collections and video streams. Notable publications include "Building Rome on a Cloudless Day" (European Conference on Computer Vision, 2010), which demonstrated efficient dense reconstruction of urban scenes from millions of images; "Building Rome in a Day" (International Conference on Computer Vision, 2009); "Detailed Real-Time Urban 3D Reconstruction from Video" (International Journal of Computer Vision, 2008); and "USAC: A Universal Framework for Random Sample Consensus" (IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013). These works have garnered over 30,000 citations according to Google Scholar, underscoring his influence in the field. Frahm has advised numerous doctoral students, including those whose dissertations focused on topics in 3D reconstruction and dynamic scenes. His collaborative efforts have resulted in best paper awards, such as the MIUA Best Paper Award for "Surface Registration in the Presence of Missing Patches."

Professional Email: jmf@cs.unc.edu