Workshop on

Generative-Model Based Vision:

GMBV 2004

Washington DC, USA
2 July 2004


Preliminary Program

08:30 Opening remarks
08:40 Session I
Modeling Images of Natural 3D Surfaces: Overview and Potential Applications
Andre Jalobeanu, Frank O. Kuehnel, and John C. Stutz
(NASA Ames Research Center)
Information Scaling Laws in Natural Scenes
Cheng-en Guo, Ying Nian Wu, and Song Chun Zhu
(University of California, Los Angeles)
09:20 Question period for Session I
09:30 Session II
A Probabilistic Approach to Linear Subspace Fitting for Computer Vision Problems
Takayuki Okatani
(Tohoku University)
Wedgelet Enhanced Appearance Models
S. Darkner, R. Larsen, M. B. Stegmann, and B. K. Ersbøll
(Technical University of Denmark)
10:10 Question period for Session II
10:20 Coffee/tea
10:40 Session III: Object Recognition A
Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories
Li Fei-Fei, Rob Fergus*, and Pietro Perona
(California Institute of Technology and *University of Oxford)
Object Class Recognition with Many Local Features.
Scott Helmer and David Lowe
(University of British Columbia)
11:20 Question period for Session III
11:30 Session IV: Object Recognition B
Interpretation of Complex Scenes Using Generative Dynamic-Structure Models
Sinisa Todorovic and Michael C. Nechyba
(University of Florida)
Fast Unsupervised Greedy Learning of Multiple Objects and Parts from Video
Michalis K. Titsias and Christopher K. I. Williams
(University of Edinburgh)
12:10 Question period for Session IV
12:20 Lunch
13:20 Session V: Pose Refinement and Tracking
Visual Hand Tracking Using Nonparametric Belief Propagation
E. B. Sudderth, M. I. Mandel, W. T. Freeman, and A. S. Willsky
(Massachusetts Institute of Technology)
Shape Matching with Belief Propagation: Using Dynamic Quantization to Accomodate Occlusion and Clutter
James Coughlan and Huiying Shen
(Smith-Kettlewell Institute)
3D Tracking of Morphable Objects Using Conditionally Gaussian Nonlinear Filters
Tim K. Marks, John Hershey, J. Cooper Roddey, and Javier R. Movellan
(University of California, San Diego)
14:20 Question period for Session V
14:35 Coffee/tea
14:55 Session VI: Models for Image Sequences A
A Probabilistic Approach to Optical Flow based Super-Resolution
Rik Fransens, Christoph Strecha, and Luc Van Gool
(University of Leuven)
Nonlinear Generative Model for Dynamic Shape and Dynamic Appearance
Ahmed Elgammal
(Rutgers University)
15:35 Question period for Session VI
15:50 Session VII: Models for Image Sequences B
A Statistical Approach to Large Deformation Diffeomorphisms
Bo Markussen
(University of Copenhagen)
3D Periodic Human Motion Reconstruction from 2D Motion Sequences
Zonghua Zhang and Nikolaus F. Troje
(Queen's University)
16:30 Question period for Session VII
16:45 Session VIII: Stereo Matching
Bayesian Stereo Matching
Li Cheng and Terry Caelli
(University of Alberta)
Stochastic Diffusion for Correspondence Estimation and Object Segmentation
Sang Hwa Lee, Nam Ik Cho, and Jong-il Park
(Seoul National University)
17:25 Question period for Session VIII
17:40 General Discussion
18:00 Closing