No-reference quality assessment for video streams based on China Mobile Multimedia Broadcast

Abstract

This paper investigates the perceptual quality of multimedia affected by compression and implements a system of quality assessment for compressed video over CMMB (China Mobile Multimedia Broadcasting standards). It can evaluate the video quality of CMMB video streams and get a good corresponding with the subjective assessments results. We examine the impact of several factors on the perceptual quality, and extract various feature parameters directly from the streams, then select representative feature parameters as the inputs of our model. Finally we propose a no-reference metric that measures the overall quality degradation due to lossy compression. The experimental results show that our model can achieve a good performance for the streaming video quality prediction over CMMB. In addition, since our model does not require the source picture, or fully decoding process and has a low complexity, it is also suitable for many other real-time video applications.