Reference-Free Video Quality Metric for Mobile Streaming Applications

Mobile multimedia streaming applications are becoming more and more popular although the perceptual video quality for such low bit rates, frame rates and resolutions is limited. Depending on the content character of a video sequence, the compression settings maximizing the subjective perceptual quality also differ. Aim of this work is to find a low-complexity referencefree estimation of visual perceptual quality, based on a combination of a possibly small set of the most important objective parameters compression settings and content features. To achieve this, various objective parameters are mapped on mean opinion score (MOS), obtained by an extensive survey. Finally, an estimation is chosen having the best correlation with the measured data set. The so obtained quality metric does not require any knowledge about the original sequence. Its correlation with the data set is as good as if more complex human vision based estimation was applied.