Computer Science Colloquia: Sun, Nov 20 at 10:30 - Michael Chertok
Niv Buchbinder
Dear all,
Next week we are happy to have Michael Chertok from School of Engineering, Bar-Ilan University.
Note that we moved forward the talk to 10:30!
You can view the upcoming talks of the seminar at:
See you,
Niv Buchbinder.
%%%%%%%%%%%%%%%%%%%%%%
Computer Science Colloquia - Open University of Israel
%%%%%%%%%%%%%%%%%%%%%%
Time: Sunday, November 20, at 10:30-11:30
Place: Visitor center 194, Open University.
Speaker: Michael Chertok, School of Engineering, Bar-Ilan Univsity.
-------
Title: Spectral Symmetry Analysis
-----
Abstract:
Spectral relaxation was shown to provide an efficient approach for solving a gamut of computational problems, ranging from data mining to image registration. We show that in the context of graph matching, spectral relaxation can be applied to the detection and analysis of symmetries in n-dimensions. First, we cast symmetry detection of a set of points in Rn as the self-alignment of the set to itself. Thus, by representing an object by a set of points S \in Rn, symmetry is manifested by multiple self-alignments. Secondly, we formulate the alignment problem as a quadratic binary optimization problem, solved efficiently via spectral relaxation. Thus, each eigenvalue corresponds to a potential self-alignment, and eigenvalues with multiplicity greater than one correspond to symmetric self- alignments. The corresponding eigenvectors reveal the point alignment and pave the way for further analysis of the recovered symmetry. We apply our approach to image analysis, by using local features to represent each image as a set of points. Last, we improve the scheme’s robustness by inducing geometrical constraints on the spectral analysis results. Our approach is verified by extensive experiments and was applied to two and three dimensional synthetic and real life images.