Joint image registration and reconstruction from compressed multi-view measurements
We present a method for joint reconstruction of a set of images representing a given scene from few multi-view measurements obtained by compressed sensing. We model the correlation between measurements using global geometric transformations represented by few parameters. Then, we propose an algorithm able to jointly estimate these transformation parameters and the observed images from the available measurements. This method is also robust to occlusions appearing in the scene. The reconstruction algorithm minimizes a non-convex functional and generates a sequence of estimates converging to a critical point of this functional. Finally, we demonstrate the efficiency of the proposed method using numerical simulations.