Multimedia has become an integral component of the Internet and modern digital experiences, playing a vital role in our everyday lives. The majority of content that we consume daily combines both audio and video modalities. This emphasizes the demand for an effective audio-visual quality assessment. However, current state-of-the-art audio-visual quality assessment methods often require a high computational cost and execution time. Moreover, they ignore important factors, such as the signal’s context, that influence what users expect in terms of quality across different circumstances. In this study, we therefore introduce an approach that addresses these problems. Inspired by the complementary nature of audio and vision, our methods leverage audio cues to selectively amplify visually informative regions, allowing the model to capture the cross-modal contextual dependencies in a computationally efficient manner. Experimental results show that our proposed model significantly reduces the computational complexity by more than 80% of both parameter size and runtime while still providing a favorable performance compared to state-of-the-art methods. The source code of this work will be made publicly available.

As Virtual Reality (VR) systems increasingly incorporate biometric sensing and adaptive interaction techniques to enhance Quality of Experience (QoE), they also raise significant concerns regarding user privacy. While these innovations promise more immersive and personalised experiences, they often rely on the collection of highly sensitive data—such as physiological signals or behavioural patterns—without sufficient consideration of what this implies for privacy. In this position paper, we argue that privacy must be treated as a fundamental concern in QoE research, particularly as VR technologies transition from controlled experimental settings to real-world applications. We begin by outlining the current disconnect between QoE and privacy research. We then propose a research agenda aimed at reconciling QoE and privacy in VR, structured around the integration of privacy in all phases of the research process, from the research design to the analysis and dissemination of results. By embedding privacy into QoE research, we aim to foster the development of VR systems that are not only immersive and engaging, but also ethically responsible and socially sustainable.

This paper discusses the problem of ecological va-lidity of Quality of Experience experiments in a broader context of internal and external validity. We argue that the issue of trade-off between control in the experiment and generalization of the results requires diverse experimental protocols. To enable this process comparability of experiments has to be guaran-teed. To improve comparability and communication between researchers, we propose using theoretical models as tools for clearly communicating the assumptions underlying subjective experiments. These models describe the relationships between variables and enable comparability of research. We demonstrate how the theoretical model of video QoE can be used for the comparison of three independent QoE studies. In conclusion, we stress that ecological validity is a mean to an extended external validity. We suggest that new experimental protocols that increase the number of Influence Factors (IFs) measured have to provide comparability with other experiments. Theoretical models can facilitate this process and stimulate higher comparability between studies in QoE subdomains. A community effort is needed to adjust the theoretical model to new use cases, particularly with new immersive multimedia. This paper inspires further research on ecological validity in QoE and encourages the adoption of theoretical models for more comparable research designs.

In traditional subjective video quality experiments, the presented sequences are short and quality ratings are based on a single interaction with a service (i.e., one session). However, in real-life scenarios, users interact with a video service for a longer period of time. If decisions are made, such as to abandon a service, they are formulated based on longitudinal multi-episodic interaction. Therefore, it is important to better understand how quality is perceived in a longer interaction and how quality perception is linked to behavioral implications. My PhD work encompasses a longitudinal study of users’ interactions with a video service using a mobile device. In our study, which consists of six phases, we use different study designs to investigate how users perceive quality in a more ecologically valid setting. The study is carried out using a previously validated setup, which consists of compression software and a mobile application.

Most user studies in the QoE multimedia domain are done by asking users about quality. This approach has advantages: it obtains many answers and reduces variance by repeated measurements. However, the results obtained in this context may be different from those obtained in the real application, since quality is not asked about this often in everyday life. It is more natural is to focus on user behavior. The proposed PhD focuses on a method for performing experiments based on observations of a participant’s behavior. We address two main challenges that exist in any new experiment design: how to calculate the interval validity of the proposed method and how to analyze the obtained data. The data analysis we propose is based on psychometric functions. We propose two different experiments, one of which is already ongoing.

This tools track paper presents the design of a questionnaire methodology to assess the participants experience of virtual conferences. This survey approach consists of a preconference questionnaire assessing participation goals and expectations and a post-conference questionnaire assessing the actual participation and related experiences. It enables a data-driven investigation of participants' expectations, goals, attitudes, actual experiences, and general feedback about virtual conferences. As such, it can help to better understand how virtual conference experiences can be improved in the future and how the virtual format can become a more attractive alternative, also in non-pandemic times. The questionnaire was used at three conferences and two workshops. Despite a missing validation, we released it early to foster research on virtual conferences.

During the Covid-19 pandemic, the use of videoconferencing for medical consultations between patients and general practitioners (GPs) has strongly increased. The aim of this work-in-progress paper is to investigate which factors (technical and non-technical) influence patients' and doctor's Quality of Experience with video consultations (“tele-consultations”) and how. We report on findings from an online questionnaire (N=109) and an (ongoing) semi-structured interview study targeting GPs (N=4), highlighting the importance of both technical and nontechnical factors.

Today's software offers diverse functionalities that need to be made accessible to humans in an easy to understand and quick to learn way. This is not a new challenge and already well addressed in usability research. Yet, similar challenges arise in crowdsourcing tasks. While task interfaces are often less complicated than complete software products, crowdsourcing workers have only a minimal amount of time to familiarize them with the task interface. Additionally, due to the repetitiveness of the tasks, workers are required to work with this interface to solve a few hundred tasks. Unfortunately, often little attention is paid to optimize the task interfaces. Even if recent studies have shown evidence of the negative effects of poorly designed interfaces on the work performance, there is less knowledge about the importance of the usability from the worker's point of view and their personal preferences. This work aims to fill this gap in two steps. First, we analyze the relevance of good usability and interface design in relation to other task properties like payment or joyfulness by conducting a survey on two popular crowdsourcing platforms. Second, we identify interface properties that are of importance for the crowdsourcing workers and discuss both consenting and dissenting opinions of different worker groups. The results show that all workers place an essential role in the usability of a task interface when selecting tasks, but do not agree on coherent design preferences within and between the platforms.

Some of the most important aspects for content creators and service providers are the content appeal and the time consumed by the end users on a particular application or service. Gender and age can influence the Quality of Experience (QoE) ratings of multimedia based on the nature of the shown content, yet few studies have quantized this notion. In this paper, we zoom in on the influence of gender on user ratings in a video QoE study (N=89) with packet loss as the main system influence factor. We have analyzed the impact of gender on QoE subjective ratings both as a standalone influence factor and in coherence of temporal traits like the frequency of watching online content. We have observed significant trends to highlight the importance of systematically checking and reporting on the impact of basic human factors, such as gender in relation to quality perception with respect to different types of content.

We address the challenge faced by service providers in monitoring Quality of Experience (QoE) related metrics for WebRTC-based audiovisual communication services. By extracting features from various application-layer performance statistics, we explore the potential of using machine learning (ML) models to estimate perceivable quality impairments and to identify root causes. We argue that such performance-related data can be valuable and informative from a QoE assessment point of view, by allowing to identify the party/parties in a call that is/are experiencing quality impairments, and to trace the origins and causes of the problem. The paper includes case studies of multi-party videoconferencing that are established in a laboratory environment and exposed to various network disturbances and CPU limitations. Our results show that perceivable quality impairments in terms of video blockiness and audio distortions may be estimated with a high level of accuracy, thus proving the potential of exploiting ML models for automated QoE-driven monitoring and estimation of WebRTC performance.

Young children represent an important, yet under-explored user segment in QoE research. This is partly due to the limitations associated with traditional QoE assessment methods. As a first step towards addressing this non-traditional user segment, this paper shares selected findings from a mixed-method study on the use of video chat and potential influence of technical influence factors on children's QoE. An exploratory survey with parents (N=87) was combined with home observations (N=7) and focus groups with children (N=21). The results provide further insights into how and why video chat is used with young children, and initial insights into how children react to technical impairments, e.g., by displaying emotional or behavioral indications of poor QoE. Further, we reflect on a number of methodological challenges that need to be addressed in future QoE-studies with children.

Enabling high Quality of Experience (QoE) with video-conferencing solutions based on Web Real-Time Communication (WebRTC) protocols anywhere, anytime, is challenging and triggers the exploration of new ways to gain QoE insights. In this paper, we share initial observations from a within-subjects experiment (N = 22) in which 2-party WebRTC-based audiovisual conversations took place under varying technical conditions. We collected self-report data, peripheral physiological data and application-level performance statistics with the overall aim of exploring their usefulness and compatibility for assessing QoE in the context of WebRTC. Our preliminary findings indicate that the varying quality is well-reflected in the self-reported overall quality and annoyance. However, this might not apply when considering the emotional valence ratings and ECG data, implying that other factors may play a role here.

In an increasingly connected world, users access personal or shared data, stored "in the cloud" (e.g., Dropbox, Skydrive, iCloud) with multiple devices. Despite the popularity of cloud storage services, little work has focused on investigating cloud storage users' Quality of Experience (QoE), in particular on mobile devices. Moreover, it is not clear how users' context might affect QoE. We conducted an online survey with 349 cloud service users to gain insight into their usage and affordances. In a 2-week follow-up study, we monitored mobile cloud service usage on tablets and smartphones, in real-time using a mobile-based Experience Sampling Method (ESM) questionnaire. We collected 156 responses on in-situ context of use for Dropbox on mobile devices. We provide insights for future QoE-aware cloud services by highlighting the most important mobile contextual factors (e.g., connectivity, location, social, device), and how they affect users' experiences while using such services on their mobile devices.

Users in todaypsilas mobile ICT environment are confronted with more and more innovations and an ever increasing technical quality, which makes them more demanding and harder to please. It is often hard to measure and to predict the user experience during service consumption. This is nevertheless a very important dimension that should be taken into account while developing applications or frameworks. In this paper we demonstrate a software tool that is integrated in a wireless living lab environment in order to validate and quantify actual user experience. The methodology to assess the user experience combines both technological and social assets. User experience of a Wineguide application on a PDA is related to signal strength, monitored during usage of the applications. Higher signal strengths correspond with a better experience (e.g. speed). Finally, difference in the experience among users will be discussed.