Statistical and Computational Theories of Vision
-- Modeling, Learning, Computing, and Sampling
Fort Collins, Colorado, June 22, 1999
(The day before CVPR99
Selected papers will be published in a special issue of
Int'l Journal of Computer Vision .
Goals and Scope
This workshop will provide an international forum for discussing recent
advances in the statistical and computational theories of vision. It will
brings together highly respected researchers from computer vision, mathematics,
statistics, information theory, and psychology to address key issues in
Bayesian image analysis, such as
Advanced theories for statistical modeling,
Advanced theories for visual learning,
Advanced techniques for efficient computation and search,
Advanced techniques for Markov chain Monte Carlo Sampling.
Advanced theory for performce bound.
Original papers are solicited on topics including, but are not
Analyzing the statistics of natural imagery and patterns.
Statistical modeling of natural patterns (textures, shapes, biometrics...)
Statistical modeling of clutter and sensor noise.
Statistical modeling of scenes.
Visual learning, model estimation and model selection.
Information theoretic principles in computer vision, such as
max. entropy, max. mutual info., minimax entropy,...
Markov chain Monte Carlo sampling methods, such as
Importance sampling, stochastic diffusion, jump-diffusion, multi-try metropolis, IWIW,...
Algorithms for efficient computation and search, such as
graph partitioning, 20-questions, Markov tree, dynamic search,...
Performance bounds, order numbers, measure of scene complexity.
Scene representation and graph matching.
Applications: image segmentation, perceptual organization, object
recognition, motion analysis and tracking, medical images, ...
Organizing Committee :
Alan Yuille, David Mumford
Program Committee :
| Edward Adelson,
Paper submission deadline: March 1, 1999
Notification of acceptance: May 1, 1999
Submission of Papers:
Prospective authors are invited to submit papers to
the following address. The submission includes One cover page with title, author names and address.
Four copies of manuscript, no longer than 25 pages including
figures and references.
Please keep the author's name blind to the reviewers in the manuscripts.
|Prof. Song-Chun Zhu
Dept. of Computer and Information Science
The Ohio State University
2015 Neil Avenue
Columbus, OH 43210
Remark: this workshop is aimed at very high quality
presentation and publication.
Selected papers will be published in a special issue of Int'l Journal of Computer Vision.