Generalized Camera Array Model for Standard Plenoptic Cameras

Abstract

Plenoptic cameras discriminate the contribution of each ray emanating from a particular point by placing a microlens array between the main lens and the image sensor. The collection of rays captured by these cameras can represent the physical microlens camera array or can be rearranged to represent a virtual camera array with a very narrow baseline (viewpoint camera array). In this work, we extend the common camera arrays considered for standard plenoptic cameras (SPCs) and define the geometry associated with the different virtual camera arrays that can be obtained by shearing the lightfield (LF). This geometry is validated using a publicly available calibration dataset and calibration toolbox. The results show that the geometry proposed is capable of describing the multiple viewpoint and microlens camera arrays.