A New Approach for Estimating Depth by Fusing Stereo and Defocus Information

Conference paper


Ioana Gheta
Christian Frese
Michael Heizmann
Jürgen Beyerer


37. Jahrestagung der Gesellschaft für Informatik e.V. (GI): INFORMATIK 2007 - Informatik trifft Logistik, vol. 1, 2007.




Workshop Image Fusion, 37. Jahrestagung der Gesellschaft für Informatik e.V. (GI), Bremen, September 24 - 27, 2007

Several algorithms are common for estimating depth from stereo series, but many of them have difficulties when determining depth of objects having periodical structure. This contribution proposes a method to overcome the impediments by using defocus as additional information. The algorithm fuses depth from stereo and depth from defocused edges by analyzing and evaluating image series with simultaneously varied camera and focus positions. The problem is formulated by a comprehensive notation using energy functionals, which can be solved e. g. by applying graph cuts minimization.