Video Quality Assessment Considering the Features of the Human Visual System

Nowadays, numerous video compression quality assessment metrics are available. Some of these “objective” and only tangentially represent how a human observer rates quality. On the other hand, models visual system have been shown to be effective at describing spatial coding. In this work we propose new metric which extends peak signal noise ratio with features measured using modern LCD screens. We also analyse current visibility early compare commonly used containing data modelling perception. examine Pearson’s linear correlation coefficient various subjective scores on videos from publicly available Netflix set. Of tested, our proposed is found most stable high performance in predicting