A general parametric model for perceptual video quality estimation

In this paper a general parametric model is proposed which provides estimation for the perceived quality of video, coded with different codecs, at any bit rate and display format. The proposed model takes into account video content, using an objective estimation of the spatial-temporal activity, based on the average SAD (Sum of Absolute Differences) for the clip. Studies were made for more than 2000 processed video clips, coded in MPEG-2 and H.264/AVC, in bit rate ranges from 25 kb/s to 12 Mb/s, in SD, VGA, CIF and QCIF display formats. The results shows that the proposed model fits very well to the perceived video quality, in any combination of codec, bit rate and display format.