SIG-09 Program:

8:30 - 8:35 Song-Chun Zhu (University of California, Los Angeles)

Welcome and Opening Remarks

8:35 - 8:50 Sinisa Todorovic (Oregon State University)

Brief Review of Grammars (Slides)

8:50 - 9:30 Stuart Geman (Brown University)

Keynote talk: Generative Hierarchical Models for Image Analysis (Extended Abstract, Slides)

9:30 - 10:10 Ales Leonardis (University of Ljubljana)

Keynote talk: Learning a Hierarchical Compositional Representation of Multiple Object Classes (Extended Abstract, Slides)

Sanja Fidler, Marko Boben, and Ales Leonardis (University of Ljubljana)

A bottom-up and top-down optimization framework for learning a compositional hierarchy of object classes (Extended Abstract)

10:10 - 10:30

Coffee break

10:30 - 11:05 David Mumford (Brown University)

Keynote talk: Is There a General Structure for Grammars? (Extended Abstract, Slides)

11:05 - 12:00 Alan Yuille (University of California, Los Angeles)

Keynote talk: Recursive Compositional Models: Representation, Learning, and Inference (Extended Abstract)
(joint work with Long Leo Zhu)

Iasonas Kokkinos and Alan Yuille (Ecole Centrale de Paris and University of California, Los Angeles)

Inference and Learning with Hierarchical Compositional Models (Extended Abstract)

12:00 - 1:30

Lunch break

1:30 - 1:50 Vinay Shet, Maneesh Singh, Claus Bahlmann, and Visvanathan Ramesh (Siemens Corporate Research, Princeton, USA)

Predicate Logic based Image Grammars for Complex Pattern Recognition (Extended Abstract)

1:50 - 2:20 Benjamin Yao, Xiong Yang, and Tianfu Wu (Lotus Hill Institute and University of California, Los Angeles)

Image Parsing with Stochastic Grammar: The Lotus Hill Dataset and Inference Scheme (Extended Abstract)

2:20 - 3:05 Narendra Ahuja (University of Illinois at Urbana-Champaign)

Keynote talk: A Syntax for Image Understanding (Extended Abstract, Slides)

3:05 - 3:20

Coffee break

3:20 - 3:40 M. S. Ryoo and J. K. Aggarwal (University of Texas at Austin)

Stochastic Representation and Recognition of High-level Group Activities (Extended Abstract)

3:40 - 4:25 Sven Dickinson (University of Toronto)

Keynote talk: Beyond One-to-One Feature Correspondence: The Need for Many-to-Many Matching and Image Abstraction (Extended Abstract, Slides)

4:25 - 5:00

Panel discussion