Reconstructing Dense Light Field from a Multi-Focus Images Array for Virtual View Synthesis

Abstract

This paper presents a novel method for synthesizing a virtual view from two sets of differently focused images taken by an aperture camera array for a scene of two approximately constant depths. The proposed method consists of two steps. The first step is a view interpolation to reconstruct an all in-focus dense light field of the scene. The second step is to synthesize a virtual view from the reconstructed dense light field by using light field rendering technique. The view interpolation in the first step can be achieved simply with linear filters that are designed to shift different object regions separately without estimating the depth map of the scene. The proposed method can effectively create a dense array of pin-hole cameras (i.e., all-focused images) so that the virtual view can be synthesized with higher quality.