Evaluating the effectiveness of image quality metrics in a light field scenario
Giuliano Arru, Federica Battisti, Marco Carli · Electronic Imaging · 2019
In this contribution, an objective metric for quality evaluation of light field images is presented. The method is based on the exploitation of the depth information of a scene, that is captured with high accuracy by the light field imaging system. The depth map is estimated both from the original and impaired light field data. Then, a similarity measure is applied, and a mapping is performed to link the depth distortion with the perceived quality. Experimental test performed by comparing state-of-art metrics with the proposed one, demonstrate the effectiveness of the proposed metric.