Omni 3D project
300R: Mathematics of 3D image synthesis
News
by Oliver Knill
Office: SciCtr 434
Email: knill@math.harvard.edu
This supervised reading and research course of Summer 2007
explored mathematical tools for
extracting information about a scene from a sequence of pictures.
These inverse problems involve different mathematical areas like
projective geometry and algorithms from statistical optimization,
linear algebra, dynamic programming and image processing. This research was done
with Jose Luis Ramirez-Herran. We focussed primarily
on the structure from motion problem in computer vison, which is the inverse problem
to reconstruct camera and object from image data. Here are three papers, we wrote:
|
| This reading and research course Maths300R was sponsored by the Harvard summer school in the Summer of 2007. |
| Update of November 25, 2008: the result on Ullman's theorem has been obtained earlier: by Huang, T. S. and Lee, C. H. 1989. Motion and Structure from Orthographic Projections. IEEE Trans. Pattern Anal. Mach. Intell. 11, 5 (May. 1989), 536-540. . |