Video Quality Prediction based on the Spatial and Temporal Classification of the Uncompressed Content

This paper deals with the notion of user satisfaction relative to the consumption of modern encoded video applications and services. Towards this, the concept of the perceived quality of service (PQoS) is introduced and exploited for quantifying purposes. The objective of the paper is to research the relationship of the spatiotemporal dynamics of the content to the deduced perceptual quality of a MPEG-based encoded video. Once this relationship has been established, it is shown how it can be further exploited towards proposing a technique for video quality prediction at a pre-encoding state.