Integrating the everyday-life context in subjective video quality experiments

Abstract

Controlled subjective quality experiments are a well-known method to make decisions on the improvement of Quality of Service (QoS) of video streams. Recently it became clear that from a point of view of the consumer or user, the Quality of Experience (QoE) is more relevant and can influence the optimal QoS as determined in a lab-setting. The measurement of QoS parameters in a lab-setting does not take into account the specific context of the practice that is under examination (e,g. watching video content). In this paper we discuss a method of contextualized subjective quality experiments as we applied in different research projects in complement to the standardized lab-experiments. The strength of this method is that video and audio quality is assessed in the real-life context of the user, i.e. his or her natural habitat in which the behavior or practice normally takes place. We provide an overview of how we applied the contextualized research approach in two cases. Next we discuss the method's strengths and weaknesses and suggest a refinement of the methodology.