No Reference Video Quality Evaluation for Multimedia Applications

Perceptual video quality evaluation plays important roles in multimedia applications, such as video teleconferencing, video on demand and Internet streaming media. Most of the traditionally objective video quality assessment attempted to quantify the predefined artifacts of the coded video sequences. In this paper, we present a new no reference perceptual video quality evaluation model based on spatial and temporal features irrespective of any predefined artifacts in Multimedia applications. The proposed spatial features are pixel distortion and edge information measurement of each frame and the temporal features measure are effective temporal information, effective frame and frame similarity measurement of the sequence. Subjective experiments results are used to verify the model, and show that the proposed model performances are reasonably close to the subjective scores.