HVS Based Full Reference Video Quality Assessment Based on Optical Flow

The use of statistical features of optical flow: mean, standard deviation of flow magnitudes and minimum eigenvalue of optical flow patch's covariance matrix, in measuring the distortion levels is well demonstrated in the state of art [1]. We hypothesize that when there is a higher random flow in both the reference video and distorted video, then the temporal annoyance level is low which result in reduced distortion scores. Based on this hypothesis, we present an HVS based full reference video quality assessment algorithm based on optical flow. The experimental results demonstrate that the proposed model improves the state of art.