CMSC 828Z::

Class Description :: Syllabus :: Reading List :: Projects :: Helpful Links ::

CMSC828Z: Description of Project 9 - Reconstruction of Super-resolution Textures and 3D Flow fields on the Surface of the Reconstructed Object

The object is observed by 33 calibrated video cameras simultaneously. The image sequences captured by the cameras show the same location on the object surface from many different views and at many different times. Using the techniques from super-resolution image reconstruction, we would like to utilize the redundancy in the images to extract highly detailed texture maps and 3D scene flow fields on the object surface. Starting pointers for this project are:

• Fundamentals of Texture Mapping and Image Warping, Paul Heckbert, Master's thesis, UCB/CSD 89/516, CS Division, U.C. Berkeley, June 1989, 86 pp
• Simon Baker and Takeo Kanade. Super resolution optical flow. Technical Report CMU-RI-TR-99-36, Robotics Institute, Carnegie Mellon University, October 1999.
• F. Bernardini, I. Martin, and H. Rushmeier. High-quality texture reconstruction. IEEE Trans. on Vis. and Comp. Graph., 7(4), October-November 2001.
• L. Wang, S. B. Kang, R. Szeliski, and H.-Y. Shum. Optimal texture map reconstruction from multiple views. In Proc. IEEE Conference on Computer Vision and Pattern Recognition, volume I, pages 347-354, 2001.