In our study, we explore the human side of the multimedia experience. The authors propose a model that assesses quality variation from three distinct levels: the network-, the mediaand the content-levels; and from two views: the technicaland the user-perspective. By facilitating parameter variation at each of the quality levels and from each of the perspectives, we were able to examine their impact on user quality perception. Results show that: a significant reduction in frame rate does not

more »

... ortionally reduce the user's understanding of the presentation, independent of technical parameters; the type of video clip significantly impacts user information assimilation, user level of enjoyment and user perception of quality; the display type impacts user information assimilation and user perception of quality. Finally, to ensure transfer of informational content, network parameter variation should be adapted; to maintain user enjoyment, video content variation should be adapted. Defining Multimedia Quality Distributed multimedia quality is not defined by a "single monotone dimension"; it is judged instead using numerous factors, which have been shown to influence user criteria concerning presentation excellence, e.g. delay or loss of frames, audio clarity, lip synchronisation during speech, as well as the general relationship between visual auditory components [2]. As a result, considerable work has been done looking at different aspects of distributed multimedia video quality at many different levels. Due to these multiple influences, the comparable examination of perceived quality becomes complex. To aid this comparison this paper extends a quality definition model first used by Wikstrand [33] that segregates quality into three discrete levels: the network-level, the media-level and content-level. Wikstrand showed that all factors that influence distributed multimedia quality can be categorised by assessing the information abstraction. The network-level concerns the transfer of data and all quality Defining the Users Perception of Distributed Multimedia Quality Gulliver & Ghinea 2 issues related to the flow of data around the network. The media-level concerns quality issues relating to the transference methods used to convert network data to perceptible media information, i.e. the video and audio media. The content-level concerns quality factors that influence how media information is perceived and understood by the end user. • The network-level is concerned with how data is communicated over the network and includes variation and measurement of parameters including: bandwidth, delay, jitter and loss. • The media-level is concerned with how the media is coded for the transport of information over the network and / or whether the user perceives the video as being of good or bad quality. Media-level parameters include: frame rate, bit rate, screen resolution, colour depth and compression techniques. • The content-level is concerned with the transfer of information and level of satisfaction between the video media and the user, i.e. level of enjoyment, ability to perform a defined task, or the user's assimilate critical information from a multimedia presentation. At each quality abstraction defined in Wikstrand's model, quality parameters can be varied, e.g. jitter at the network-level, frame rate at the media-level and finally display-type at the content-level. Similarly, at each level of the model, quality can be measured, e.g. percentage of loss at the network-level, user mean opinion score (MOS) at the media-level, and task performance at the content-level. As well as possessing three distinct information abstractions, distributed multimedia covers a range of applications, which reflects the symbiotic infotainment duality of multimedia, i.e. the ability to transfer information to the user, yet also provide the user with a level of subjective satisfaction in respect of its perceived quality. Consequently, the user perspective concerning multimedia quality should consider both how a multimedia presentation is understood by the user, yet also examine the user's level of satisfaction (both satisfaction with the perceived Quality of Sevice setting and level of enjoyment concerning the video material). As multimedia applications are ultimately produced for the education and / or enjoyment of human viewers, the user's perspective concerning the presentation quality is surely of considerable importance. Accordingly, distributed multimedia quality, in our perspective, is deemed as having two main facets: Quality of Service (QoS) and Quality of Perception (QoP). The former (QoS) characterises the technical perspective and represents the performance properties provided by multimedia technology. The latter facet (QoP) considers the user perspective, measuring the infotainment impact of the presentation. Accordingly, and