Topic This tutorial will present the state of the art in 3D reconstruction  and camera pose recovery from multiple views of static and dynamical scenes.  Applications of the techniques presented include autonomous navigation, building 3D models from video, tracking, visual surveillance, augmented reality and special effects for entertainment. The course is based on a recently published  textbook co-authored by the organizers, and includes exercises, live demonstrations and sample code of the covered techniques. The organizers prepared the similar tutorial for ICRA 2003, which was one of the best attended tutorials at the conference.  Thanks to the recent developments in the field, previous topics will be extended to include comprehensive treatment of the analysis and recontruction of the dynamic scenes and utilization of prior scene knowledge in the reconstruction. These recent additions will also be supported by new demonstrations of the applicability of the presented algorithms. The tutorial will be supported by a textbook titled "Invitation to 3D Vision: From Images to Geometric Models",  (Springer-Verlar, December 2003) that was recently published by the organizers and includes excercises and sample MATLAB code of the algorithms covered in the tutorial.   Upon completing the course, participants should possess the tools necessary to reconstruct the 3D shape of objects or scenes from video of set of photographs, calibrate cameras and recover geometric properties of dynamic scenes including multiple moving objects. Target Audience: Robotics researchers interested in using vision as a sensor for control tasks (navigation, manipulation, tracking, docking etc.) and 3-D model acquisition tasks. Robotics researchers interested in sensor networks and smart environments, medical, mobile and humanoid robotics, virtual and mixed reality.  Organizers Jana Kosecka, George Mason University,  Yi Ma, University of Illinois at Urbana-Champaign,  Shankar Sastry, University of  California at Berkeley, Stefano Soatto, University of California at Los Angeles,  R. Vidal, Johns Hopkins University Speakers same as organizers

Y. Ma	J. Kosecka	S. Sastry	S. Soatto	R. Vidal

Schedule

    8:30  - 8:45 Tutorial Overview: Geometric Techniques in Motion Understanding (S.Sastry)
    8:45  - 9:15 Rigid body motion, imaging geometry (Y. Ma) (.ps)
    9:15  - 10:00 Image primitives and correspondence (J. Kosecka) (.pdf)
   10:00 - 10:15 Coffee break
   10:15 - 11:15 Two view geometry: discrete, continuous general case, planar case (J. Kosecka) (.ps)
   11:15 - 12:00 Stratified model based reconstruction from uncalibrated views (S. Soatto) (.pdf)
   12:00 - 1:00 Lunch Break
   1:00 - 1:45  Multiview reconstruction from points and lines (Y.Ma) (.pdf)
   1:45 - 2:30  Multiple Motion Segmentation and Reconstruction Generalized Principal Component Analysis (R. Vidal) (.pdf)
   2:30 - 3:00 Coffee break
   3:00 - 3:30 Recursive filtering for motion and shape estimation (S. Soatto) (.pdf)
   3:30 - 4:00 Case study  UAV's landing (S. Sastry) (.pdf)
   4:00 - 4:30 Case study  Multiple Motion Estimation (R. Vidal) (.pdf)