Margaret Regan: „Image Reconstruction Using Numerical Algebraic Geometry“


Datum / čas
Date(s) - 30.08.

Místo konání


Margaret Regan (University of Notre Dame, Notre Dame, IN, USA)
„Image Reconstruction Using Numerical Algebraic Geometry“

Friday 2019-08-30 at 11:00
CIIRC Room B-670 (building B, floor 6)


Many problems in computer vision can be formulated using a parameterized system of polynomials which must be solved quickly and efficiently for given instances of the parameters.  We propose a new numerical algebraic geometric method to efficiently solve these  systems.  First, our new approach uses locally adaptive methods and sparse matrix calculations to solve parameterized overdetermined systems in projective space.  Examples will be provided in 2D image reconstruction to compare the new methods with traditional  approaches in numerical algebraic geometry.  Second, we propose new homotopy continuation methods for solving two minimal trifocal calibrated relative pose problems defined by point and line correspondences, which appear together, e.g., in urban scenes or  observing curves.  Simulations and comparisons will be shown using real and synthetic data to demonstrate that challenging scenes can be reconstructed where standard methods fail.

For more information on IMPACT seminars, please visit