[grads] [Sem-coll] Feb 2 Speaker Abstract
jmillham at iit.edu
Mon Feb 1 09:26:09 CST 2010
Below is the abstract for the Applied Math Speaker coming tomorrow, Feb. 2
Institute of Mathematics and Its Applications, University of Minnesota
4:40 pm E1 119
"Multi-object tracking via Bayesian techniques"
Tracking simultaneously a large collection of targets is a challenging
problem with great interest in several disciplines such as
engineering, biology, ecology, etc. In this talk, we address the
problem of dynamically estimating multiple objects within a Bayesian
filtering framework. Modeling a plethora of individual targets as a
set-valued state and adopting the random set theory as a unified
approach to multi-target tracking result in the propagation of a
multi-target Bayesian filtering distribution. However, the
multi-target filtering distribution is computationally expensive, and
thus it is approximated by a first
and a second moment, the probability hypothesis density (PHD) and the
cardinalized PHD respectively. Approximation schemes are implemented
using simulated data. This is a joint work with Ron Mahler.
See you then!
Department of Applied Mathematics
Illinois Institute of Technology
Engineering-1 Room 208
10 W. 32rd St.
Chicago IL 60616
sem-coll mailing list
sem-coll at math.iit.edu
More information about the grads