Ofer Hadar

In this paper we analyze the viewer-perceived quality of a compressed video stream, which is transmitted over a lossy IP network with QoS (quality of service) mechanism. We demonstrate that when jointly considering the impact of coding bit rate, packet loss ratio and video characteristics, there is an optimal encoding scheme, which maximizes the viewer-perceived quality. The video content, the compression and the transmission are represented in a vector X of d parameters. Based on the subjective tests, we obtain a set of observation pairs of labeled samples Pi = {Xi,Qi}, where Qi is the quality class related to the vector of input parameters Xi. We introduce a novel method to predict the optimal encoding scheme based on canonical discriminant analysis (CDA) for feature classification.