A video quality metric based on frame differencing

Assessment of visual quality is of great importance in several image and video processing applications. Various objective video quality metrics have been proposed recently to accord with the human visual perception. This paper proposes a simple and affordable objective quality metric for tracking moving objects in video streams. The proposed metric particularly takes the moving objects into account as visually important content. Foreground masks produced by frame difference-based background subtraction are incorporated as weighting function into the existing metrics. Test results show that our metrics are effective.