The impact of temporal pooling and spatio-temporal quality interaction on the prediction accuracy of video quality metrics

We investigate to what extent temporal pooling and spatio-temporal quality interaction can help to increase a video quality metric's prediction accuracy. Our results prove that a significant gain is possible. The evaluations are conducted on high-definition video material. We propose a stand-alone Gaussian weighting as an alternative pooling strategy. Furthermore, a newly designed parameter set for an IIR filter is presented. Both approaches aim at predicting a video's quality. We show that they correlate well with human perception. Moreover, we check the observation that spatial quality contributes the most to the overall video quality. Taking the results of the provided paper into account, a completely blind video quality predictor useful for real-time video quality monitoring could be established.