Experimental Assessment of Quality Metrics in Stereoscopic Imaging

Abstract

The main objective of this work is to describe the state of the art of the Stereoscopic Image Quality Assessments (SIQA) from recent years, since it covers a compendium of models from 2007 to date. Furthermore, this paper summarizes 27 algorithms from 17 authors and their possible variations giving as a result of 280 stereoscopic metrics tested. This benchmarking is not only intended for researchers on the Image Quality Evaluation but also for researchers on the field on acquisition, processing and display of stereoscopic images. To this aim, we present not only a survey on image quality metrics but also psychophysical experiments on image databases available in this field. First, we sketch a general view of the importance of Stereoscopic Imaging. Thus, we propose different classifications in order to group the state of the art of SIQA. Then, we describe the performance of 280 metrics of SIQA using LIVE 3D Image Database. Results of algorithms are evaluated with the main purpose of being a reference for researchers in the Stereoscopic image quality field who want to perform further tests and proposing future models.