ABSTRACT
This chapter reviews the history and state of the art of perceptual quality metrics for video. It discusses the factors influencing our perception of visual quality and proposes a classification of video quality metrics according to their design philosophy. The chapter reviews three different types of metrics, namely pixel-based metrics, metrics based on the psychophysical approach, and metrics based on the “engineering” approach. It discusses comparative studies of metric prediction performance. The chapter deals with a summary of the state of the art and an outlook on future developments in the field. It examines processes and properties of the human visual system to provide a common ground for the ensuing discussion of Human Visual System-based metrics. The mean squared error and the peak signal-to-noise ratio are the most popular difference metrics in image and video processing. Most psychophysical experiments focus on the threshold of visibility, whereas quality metrics and compression are often applied to clearly visible distortions.
