A naive way to create a “fake” orthophoto from a single satellite image consists in projecting the photo into a horizontal surface. This is correct when the ground is flat, and the ground is often nearly flat, however it does not give a realistic simulation of a nadir view: for example, if the initial view was slanted and the facades of some buildings were visible, these facades will still be visible in the “orthophoto”. The goal of this project is to generate synthetic orthophotos as realistic as possible, especially in urban regions. For that, a 3D model obtained from several images will be used (as explained in the TP6), and the images will be projected into this 3D model and then vertically into a geographic grid.
Supervision
Gabriele Facciolo, Carlo de Franchis, Enric Meinhardt
Bibliographic References
[1] Rau, J., Chen, N. Y., & Chen, L. C. (2002). True orthophoto generation of built-up areas using multi-view images. Photogrammetric Engineering and Remote Sensing, 68(6), 581-588.