Candidate Talk: Matching and 3D reconstruction in urban environments

Indoor and outdoor urban environments possess many regularities which can be efficiently exploited and used for general image parsing tasks, matching, or 3D dense reconstruction from multiple widely separated views. These environments exhibit often shadows, lack textured areas, or contain repetitive textures which multiply the ambiguities in standard computer vision pipelines. During my talk I will consider those specific environments and show the way we tackle the aforementioned problems.
First, I will present an approach for detecting rectilinear structures and demonstrate their use for wide baseline stereo matching, planar 3D reconstruction, and computation of geometric context. Second, I will focus on a dense stereo method utilizing properties of piecewise planarity and restricted number of plane orientations to suppress 3D reconstruction and matching ambiguities. We show how the utilization of the scene priors yields more accurate and visually plausible results in many urban scenes compared to the standard general methods.

Speaker Details

Branislav Micusik is a research scholar at the George Mason University, VA and a visitor research scholar at the Stanford University since Oct.2007. He was a postdoctoral researcher at the Vienna University of Technology, Austria, from 2004-2007. He received the Ph.D. in 2004 from the Czech Technical University in Prague. His research interests include computer vision, particularly urban scene analysis, image segmentation, omnidirectional cameras, their self-calibration and structure from motion. He co-authored a work awarded the Best Scientific Paper Prize at the British Machine Vision Conference in 2007.

Date:
Speakers:
Branislav Micusik
Affiliation:
George Mason University, VA