PROJECT TITLE :

Niqsv: A No Reference Image Quality Assessment Metric For 3d Synthesized Views - 2017

ABSTRACT:

The popularity of 3D applications, such as Free View-purpose TV (FTV) and Multi-read Video plus Depth (MVD), induces a significant demand of synthesized views. However, the quality assessment of synthesized views is terribly difficult as a result of the corresponding original views (reference views) are usually not offered at both encoder and decoder sides. During this paper, we tend to propose a replacement no-reference quality assessment model to guage the standard of 3D synthesized views, known as NIQSV (No-reference Image Quality assessment of Synthesized Views). This metric relies on the hypothesis that a sensible quality image is composed of flat areas (objects) separated by sharp edges, and the standard estimation involves solely a group of straightforward morphological operators. NIQSV integrates the distortions of all the elements, and then uses a footing image to weight the final distortions since the distortions of synthesized views mainly happen around object edges. The experimental results show that the proposed metric outperforms ancient 2D metrics and ranks among the best of dedicated 3D synthesized and full reference metrics.