The 2D-to-3D inverse projection transformation problem was finally solved in a direct manner after several unsuccessful attempts to solve it without relying on an iterative method. The problem and the direct solution are summarized in a separate document. The only remaining issue with the direct method is the possibility of division by zero (see the “Remarks” section of the summary for more information). It is unclear how often division by zero may occur. Since the iterative solver implemented and reported in the previous blog article does not have any risk of division-by-zero errors, the best approach seems to combine both, i.e., to try applying the direct method first and to switch to the iterative solver if necessary. This approach has been implemented in the latest revision 26086 of the branch. Now the bug in the 2D-to-3D inverse projection transformation is considered fully fixed. As planned in the blog article on December 28, we move on to the clipping issue in the next weeks with the aim of further improving the robustness of the view map creation.
January 18, 2010
Weekly Update January 11-17