Prof. Michael Bronstein
University of Lugano (USI)
and Intel Perceptual Computing Group
Monday, November 4th 2013, 11:00am, in BC 229
Dense correspondence between 3D shapes using sparse coding
In this talk, I will present a novel sparse modeling approach to non-rigid shape matching using only the ability to detect repeatable regions. As the input to our algorithm, we are given only two sets of regions in two shapes; no descriptors are provided so the correspondence between the regions is not know, nor we know how many regions correspond in the two shapes. I will show that even with such scarce information, it is possible to establish very accurate correspondence between the shapes by posing it as a problem of permuted sparse/structured coding. In the second part of the talk, I will show how this approach can be extended to non-isometric shapes using coupled quasi-harmonic bases constructed using joint approximate diagonalization of Laplacian operators.
Michael Bronstein received the Ph.D. with distinction from the Department of Computer Science, Technion in 2007. In 2010, he joined the Institute of Computational Science in the Faculty of Informatics at the University of Lugano (USI), Switzerland. Prior to joining USI, Michael held a visiting appointment at Stanford university. His main research interests are theoretical and computational methods in spectral and metric geometry and their application to problems in computer vision, pattern recognition, shape analysis, computer graphics, image processing, and machine learning. Michael has authored over 70 publications in leading journals and conferences, over 20 patents, and the book "Numerical geometry of non-rigid shapes" (Springer, 2008). Highlights of his research were featured in CNN, SIAM News, and Wired. His research was recognized by numerous awards, including the Kasher prize (2002), Thomas Schwartz award (2002), Hershel Rich Technion Innovation award (2003), the Copper Mountain Conference on Multigrid Methods Best Paper award (2005), the Adams Fellowship (2005), and the EUROGRAPHICS Service Award (2012). In 2012, he won the highly competitive ERC Starting Grant. Besides academic work, Michael is actively involved in industrial applications, technology transfer, and consulting to technological companies. His track record includes developing and licensing algorithms for large-scale video analysis applications at the Silicon Valley start-up company Novafora (2004-2009 as co-founder and VP of video technology) and developing coded-light 3D camera based on his patents at the Israeli start-up Invision (2009-2012 as one of the principal technologists). Following the multi-million acquisition of Invision by Intel in 2012, Prof. Bronstein currently also serves as Research Scientist at Intel.