****************************************************************
                C A L L   F O R   P A P E R S
****************************************************************

                IEEE International Workshop on
                  BEYOND MULTIVIEW GEOMETRY:
Robust Estimation and Organization of Shapes from Multiple Cues

                  in conjunction with CVPR
                 June 22, 2007, Minneapolis, MN

http://www.cs.stevens.edu/~bmgcvpr/

       Geometry-based methods such as stereo and structure from
motion have matured over the last decade. Commercial systems now
exist that reliably estimate camera calibration and motion from
movies. However, these methods have had less success at 3D shape
estimation and the organization of 3D data into coherent shapes
- except for tightly constrained problems ( e.g., architectural
scenes or multi-camera calibrated stereo). This workshop will
focus on how - and to what extent - these methods' limitations
can be transcended. Can we obtain high accuracy, dense, and robust
shape/object estimates on fairly general scenes? No single method
will suffice for this, nor will active methods such as laser
scanning. To estimate and organize accurate shapes for general
scenes, it will be necessary to combine geometry with cues such
as texture, specularity and focus, to estimate and exploit
illumination and reflectance models, to organize shapes based on
differential geometry and topology, and to explore new sensing
methodologies. The workshop will emphasize methods that achieve
robustness on real application areas through exploiting such
techniques, relying only on relatively general domain constraints.

Topics of Interest (include but are not limited to):
    * Combining geometry and reflectance/illumination information
    * Multi-cue registration
    * Organizing 3D surfaces from unorganized images.
    * Exploiting diff. geometry, topology etc. to organize shape
    * Illumination estimation and its use in shape estimation
    * Robust sensor fusion
    * High level shape representations
    * Novel sensors design which contribute in multiple cue analysis

Important Deadlines:
   Full Paper Submission:      March 19, 2007
   Notification of Acceptance: April 18, 2007
   Final Papers Due:           April 27, 2007

Paper Submission and Other Details:
   http://www.cs.stevens.edu/~bmgcvpr/

Organizing Committee:
   John Oliensis, Stevens Insitute of Technology, USA
   Elli Angelopoulou, University of Erlangen-Nuremberg, Germany
   George Kamberov, Stevens Insitute of Technology, USA
   Quynh Dinh, Stevens Insitute of Technology, USA

Confirmed Program Committee Members:
   Roberto Cipolla, University of Cambridge, UK
   Tamal Dey, Ohio State University, USA
   Paolo Favaro, Heriot-Watt University, UK
   Martial Hebert, Carnegie Mellon University, USA
   Aaron Hertzmann, University of Toronto, Canada
   Joachim Hornegger, University of Erlangen-Nuremberg, Germany
   Gerda Kamberova, Hofstra University, USA
   Kiriakos Kutulakos, University of Toronto, Canada
   Michael Langer, McGill University, Canada
   Sang Wook Lee, Sogang University, Korea
   Tomas Pajdla, Czech Technical University, Czech Republic
   Szymon Rusinkiewicz, Princeton University, USA
   Dimitris Samaras, Stony Brook University, USA
   Radim Sara, Czech Technical University, Czech Republic
   Yoav Schechner, Technion - Israel Institute of Technology, Israel
   Gregory Slabaugh, Siemens Corporate Research, USA
   Stefano Soatto, University of California at Los Angeles, USA
   Gabriel Taubin, Brown University, USA