Kitware and collaborators at the University of North Carolina, Chapel Hill; the University of Salzburg; and the University of California, San Diego, are pleased to announce that the collaborative paper “Geodesic Regression on the Grassmannian” was accepted for presentation at the European Conference on Computer Vision (ECCV) 2014. ECCV is one of the leading conferences for computer vision research, and is being held from September 6 to September 12, 2014, in Zürich, Switzerland.
The paper was co-authored by Yi Hong1, Roland Kwitt2, Nikhil Singh1, Brad Davis3, Nuno Vasconcelos4, and Marc Niethammer1,5, and it details a theory for Grassmanian geodesic regression (GGR) that extends linear regression to the Grassmannian. This work addresses the challenge of regressing data points on the Grassmann manifold over a scalar-valued variable. In addition, GGR offers a compact representation of the complete geodesic path and brings to light the possibility of statistical analysis on Grassmannian geodesics.
This research was tested on several vision challenges to demonstrate its applicability, namely the prediction of traffic speed and crowd counts from dynamical system models of surveillance videos and the modeling of aging trends in human brain structures using an affine-invariant shape representation.
To learn more about the research presented in this paper or how Kitware can help you overcome your toughest computer vision challenges through state-of-the-art software solutions, please call (518) 371-3971 or e-mail email@example.com.
1 Department of Computer Science, UNC Chapel Hill, NC, United States
2 Department of Computer Science, Univ. of Salzburg, Austria
3 Kitware Inc., Carrboro, NC, United States
4 Statistical and Visual Computing Lab, UCSD, CA, United States
5 Biomedical Research Imaging Center, UNC Chapel Hill, NC, United States