Understanding Video Quality and its use in Feedback Control

Abstract

There has been an increased interest in adaptive video quality control and dynamically adjusting the output video bit rate based on the status of the network. However, network-level performance parameters cannot accurately reflect the video quality perceived by the end users. Our goal is to investigate an adaptive perceptual video quality control mechanism based on an application-level perceptual video quality scheme. In this paper we investigate perceptual objective quality assessment technologies and use them to exploit the relationships between perceptual video quality, output bit rate, and quantization scales of video encoders. We also implemented a real time Video over IP (VIP) network application that uses a feedback channel to relay measurements, taken at the end user, to the source side, to enable the calculation of the perceptual video quality degradation caused by IP packet loss. Using this experimental setup, we were able to investigate appropriate rules for adaptive perceptual video quality control based on an application-level perceptual video quality scheme. 1.