Special session 5


Interactive Applications and Games


Steven Schmidt, Florian Metzger, and Matthias Hirth

Motivation and objectives

For many years, users mainly consumed multimedia content passively, e.g., by watching videos or listening to audio content. Therefore, even today most sophisticated user models for audio-visual (A/V) material only consider technical parameters that influence the quality of the content or the playback behavior. However, multimedia systems are rapidly changing, and users develop from passive consumers to active participants.

This active participation is most evident when it comes to video games of any kind, be it either console, PC or mobile, online or cloud, or VR and AR. In these, the users’ interactions directly control the A/V stimuli created by the system. The user involvement imposes new challenges to system designers and consequently also on the user models describing those systems. The impact of engagement on user-perceived quality and consequences of novelty effects are not yet well investigated. Furthermore, new relevant QoE aspects evolved, while issues like cybersickness must be avoided. Additionally, technical parameters such as end-to-end delay or methods to control interactive systems are paramount. However, the interactivity of applications is not only a phenomenon in the entertainment area. More and more commercial applications of VR become mature in the context of education, telemedicine, or construction processes. Lastly, also the interactivity of Web-based applications creates new challenges that have not been addressed full yet.

To overcome the challenges arising for the interactivity of games, as well as interactive VR and Web-based applications, existing test QoE test methodologies need to be refined, and entirely new techniques need to be developed. While passive A/V tests are mainly performed with pre-recorded stimuli, it remains unclear to which extent this is possible for the evaluation of the perception in interactive applications. On the contrary, it is still an open question, how to design and implement realistic emulations or testbeds for interactive applications that enable user studies at a large scale or collaborative scenarios involving multiple users. Further, existing work already indicates that a user’s familiarity with an interactive application plays an important role in the perception of impairments, however, this is not yet considered in any existing QoE model. From a conception perspective, a clear definition of ‘interactivity’ is required and categorization of interactive applications to enable structured research efforts.

These kinds of novel approaches to (subjective as well as objective) user-centric quality metrics and evaluation methods for interactive applications are the core topics we want to explore in this special session.

© QoMEX 2021