William T. Freeman
Computer Vision and Image Processing Research Interests:
- Example-based image synthesis
Image quilting for
texture synthesis and transfer.
Super-resolution
(estimating image details).
Shape-time photography (pdf file).
- Belief propagation
Bethe free energy, Kikuchi approximations, and belief
propagation algorithms:
technical report.
Generalized belief propagation:
NIPS 2000 paper.
Belief propagation in
networks with loops.
On the max-product belief propagation algorithm:
pdf file, Y. Weiss and W. T. Freeman,
IEEE Transactions on Information Theory 47:2 pages 723-735, (2001).
- Bayesian models of perception
VISTA--Vision by
Image/Scene TrAining.
Bilinear models for separating "style
and content"
Exploiting the generic viewpoint
assumption to estimate scene parameters.
Bayesian decision theory,
the maximum
local mass method, and color constancy .
Estimating 3-d body pose from
a single-camera video sequence.
- Computer vision for interactive computer graphics
Computer vision for computer games .
Television control by hand signals .
Orientation histograms for hand gesture recognition .
- Steerable Filters
Simple example showing how to
synthesize filter oriented to arbitrary angles as a linear combination
of basis filters.
Motion without movement , a method
to display motion which uses steerable filters.
Cue and contour detection using
steerable filters.
3d steerable filters .
Steerable pyramid , a
multi-scale, steerable image transform useful for image analysis and
synthesis (pointers to C code).