Automatical reconstruction of an occluded scene plane based on a camera array

Camera arrays have been used to reconstruct occluded scene planes in the recent literature. In this paper we propose a new approach to solve this task. First, by finding the local minima of the absolute difference between every pair of the input images, the images captured with a camera array are aligned according to the target plane (the plane which is occluded and needs to be reconstructed). Then, an adjustment of local details is performed which can significantly improve the quality of reconstruction. Finally, the aligned and adjusted images are averaged to get a fused image as the result of plane reconstruction. Compared with the existing methods, the proposed approach does not need a prior information about the camera array, and can be coded more easily.