Manipulation and Learning of 3D Stochastic Shape Models with Application to Archaeology
Speaker: Andrew R. Willis , Brown University
Date: December 8 2004
Time: 1:00PM to 2:00PM
Location: 32-D507
Host: Greg Shakhnarovich
Contact: Greg Shakhnarovich, gregory at csail
Relevant URL: http://hendrix.lems.brown.edu/~arwAbstract:
Automatic learning of an unknown 3D shape surface from data is an
important outstanding problem in computer vision. This lecture
discusses two new stochastic models for 3D shape and how they
contribute to shape estimation problems in archaeology : (1) a
deformable 3D mesh model for surface estimation, and (2) a system
which automatically estimates broken pots from measurements of their
pieces. In (1), a new model for stochastic deformable 3D surfaces
using a Markov Random Field is proposed and its application to
totally unconstrained 3D free-form surface sculpting and the
reconstruction of damaged artifacts is shown. In (2), a system is
described for the purpose of automatically estimating mathematical
models for archaeological vessels given 3D measurements of their
fragments commonly called sherds where it is assumed that the vessel
is symmetric about a central axis. The unique approach integrates
solutions to 4 different problems : (i) an algorithm for accurately
estimating the surface geometry of individual sherds, (ii) an
algorithm for accurately aligning assemblies of sherds, called
configurations, (iii) a Bayesian performance measure for sherd
configurations, and (iv) a performance-driven search algorithm.
Estimation of the unknown geometry of the object is implemented as
maximum likelihood estimation (MLE) where we seek to find the axially
symmetric geometry which best explains the measured fragment data.
See other events that are part of MIT Machine Vision Colloquium 2004/2005
See other events happening in December 2004
