Title: The Specialized Mappings Architecture
Authors: Romer Rosales and Stan Sclaroff
Date: March 28, 2003
Abstract:
A probabilistic, nonlinear supervised learning model is proposed:
the Specialized Mappings Architecture (SMA). The SMA employs a
set of several mapping functions that are estimated automatically
from training data. Each specialized function maps certain domains
of the input space (e.g., image features) onto the output space
(e.g., articulated body parameters). One important advantage of
the SMA is that it can model ambiguous, one-to-many mappings that
may yield multiple valid output hypotheses. Once learned, the
mapping functions generate a set of output hypotheses for a given
input via a statistical inference procedure. The SMA inference
procedure incorporates an inverse mapping or feedback function,
which enables the SMA to evaluate the likelihood of each
hypothesis. Possible feedback functions include computer graphics
rendering routines that can generate images for given hypotheses.
The SMA employs a variant of the Expectation-Maximization
algorithm for simultaneous learning of the specialized domains
along with the mapping functions, and approximate strategies for
inference. The framework is demonstrated in a computer vision
system that can estimate the articulated pose parameters of a
human body or human hands, given image silhouettes. The accuracy
and stability of the SMA are also tested using synthetic images of
human bodies and hands, where ground truth is known.