The Weizmann Institute of Science
Faculty of Mathematics and Computer Science
Vision and Robotics Seminar
1)Alfred M. Bruckstein and 2)Tal Nir
Department of Computer Science
Technion
will speak on
1) Alfred Bruckstein:Variational Image Analysis:
Do we know what to optimize for?
2) Tal Nir: Over-parameterized variational optical flow
Abstract:
We will have two short talks by two different speakers: The first talk will be
given by Freddy Bruckstein and the second by Tal Nir. Both talks are joint
work of Tal Nir, Freddy Bruckstein, and Ronny Kimmel. \medskip
1) We now have powerful numerical methods to solve variational problems for
image analysis, however the art of applying these methods is in devising the
right functionals to optimize for, and, of course, there is no procedure that
systematically yields good functionals. Tal Nir's thesis work is a good example
of new functionals that yield significant improvements in optic flow
estimation, and prompted us to revisit other variational image analysis
problems too. \medskip
2) A novel optical flow estimation process based on a spatio-temporal model
with varying coefficients multiplying a set of basis functions at each pixel is
introduced. Previous optical flow estimation methodologies did not use such an
over parameterized representation of the flow field as the problem is ill-posed
even without introducing any additional parameters: Neighborhood based methods
of the Lucas-Kanade type determine the flow at each pixel by constraining the
flow to be described by a few parameters in small neighborhoods. Modern
variational methods represent the optic flow directly via the flow field
components at each pixel. The benefit of over-parametrization becomes evident
in the smoothness term, which instead of directly penalizing for changes in the
optic flow, accumulates a cost of deviating from the assumed optic flow model.
Our proposed method is very general and the classical variational optical flow
techniques are special cases of it, when used in conjunction with constant
basis functions. Experimental results with the novel flow estimation process
yield significant improvements with respect to the best results published so
far.
The lecture will take place in the
Lecture Hall, Room 1, Ziskind Building
on Thursday, January 18, 2007
12:00 - 13:00