Traditional optical imaging faces an unavoidable trade-off between resolution and depth of field (DOF). To increase resolution, high numerical apertures (NA) are needed, but the associated large angular uncertainty results in a limited range of depths that can be put in sharp focus. Plenoptic imaging was introduced a few years ago to remedy this trade-off by reconstructing the path of light from the lens to the sensor. This is practically achieved by inserting a microlens array in the conjugate plane of the object (as created by the main lens), and moving the sensor in the conjugate plane of the main lens (as created by each microlens). The microlens array also enables the single-shot acquisition of multiple-perspective images, thus making plenoptic imaging one of the most promising techniques for 3D imaging. However, the improvement offered by standard plenoptic imaging is practical rather than fundamental: the increased DOF leads to a proportional reduction of the resolution well above the diffraction limit imposed by the lens NA, and the change of perspective is limited by the small field of view of the microlenses.
We demonstrate that this fundamental limitation can be overcome by taking advantage of the intensity correlations of light, characterizing both entangled and chaotic sources. In Correlation Plenoptic Imaging (CPI), we exploit the spatio-temporal correlation of such light sources to push plenoptic imaging to its fundamental limits of both resolution and DOF. The scheme for the theoretical and experimental demonstration of CPI is based on splitting correlated light in two arms and measuring intensity correlations between two distinct sensors. One sensor captures multiple second-order (``ghost'') images of the object, one for each pixel of the other sensor; such images are focused provided a generalized thin-lens equations is satisfied. The second sensor images the source plane, thus encoding, through the correlation measurement, information on the direction of light from the source to the object. This enables to perform the typical tasks of plenoptic imaging, such as refocusing and changing the point of view.
We experimentally prove the effectiveness of CPI and show that CPI enables a combination of resolution and DOF that is not accessible to classical imaging systems. Our results represent the theoretical and experimental basis for the effective development of the promising applications of plenoptic imaging. The plenoptic application is the first situation in which the counterintuitive properties of correlated systems are effectively used to beat intrinsic limits of state-of-the-art imaging systems.