[Brown CS Talks] Brown CS Seminar: Michael Isard in Lubrano on November 26, 2002 at 4 pm

[talks@list.cs.brown.edu]

			      CS Seminar
		  
		  The Department of Computer Science
			   BROWN UNIVERSITY

			      
			       presents

			    Michael Isard
		   
			  Microsoft Research

			
		  Tuesday, November 26, 2002 at 4 pm

	       Lubrano Conference Room (CIT 4th floor)

		Refreshments will be served at 3:45 pm
	       

	   Real-Valued Graphical Models for Computer Vision
 
			       Abstract

Probabilistic models have been adopted for many computer vision
applications, but inference in high-dimensional spaces remains
problematic.  As the state-space of a model grows, the dependencies
between the dimensions lead to an exponential growth in computation
when performing inference. Many common computer vision problems
naturally map onto the graphical model framework; the representation
is a graph where each node contains a portion of the state-space and
there is an edge between two nodes only if they are not conditionally
independent. When this graph is sparsely connected, belief propagation
algorithms can turn an exponential inference computation into one
which is linear in the size of the graph.  However belief propagation
is only applicable when the variables in the nodes are discrete-valued
or jointly represented by a single multivariate Gaussian distribution,
and this rules out many computer vision applications.

I will describe a technique to combine belief propagation with ideas
from particle filtering; the resulting algorithm performs inference on
graphs containing both cycles and continuous-valued latent variables
with general conditional probability distributions.  Such graphical
models have wide applicability in the computer vision domain; I have
tested it on example problems of low-level edge linking and locating
jointed structure in clutter.



		    Host: Professor Michael Black