Full and Short Papers

The following full and short papers were selected through a double-blind peer-review process. During the TPC meeting on March 6th in Leuven, Belgium 24% of all submitted papers were accepted, based on at least three reviews and one meta-review by an Associate Chair.

Full papers will get a 20-minute time slot (+Q&A) for presentation at the conference, short papers get a 10-minute time slot (+Q&A). The final program with the exact timing of the presentations will be made available in April.

Since the authors are preparing their camera-ready versions by April 15th, the titles and abstracts below are still subject to change.

“I’m just on my phone and they’re watching TV”: Quantifying mobile device use while watching television

Christian Holz – Yahoo Labs, Sunnyvale, CA, USA
Frank Bentley – Yahoo Labs, Sunnyvale, California, United States
Karen Church – Yahoo Labs, Sunnyvale, California, USA
Mitesh Patel – Yahoo Labs, Yahoo Labs, Sunnyvale, California, United States

Abstract: In recent years, mobile devices have become a part of our daily lives, much like television sets had over the second half of the 20th century. Increasingly, mobile devices are being used while watching a television program. We set out to understand this behavior on a minute-by-minute quantified level as well as the motivations and purposes of device use while watching television. We conducted a novel mixed-methods study inside seven households with fourteen instrumented phone and tablet devices, capturing every app launch and app use duration, correlated with the moment in the television program when it occurred. Surprisingly, we found little difference between the volume of device use during programs and commercials, but did uncover interesting patterns of device use with different genres of program as well as differences in the apps used during programs and commercials. This study sets the stage for larger-scale investigations into the details of mobile interactions while watching television. Furthermore, the novel method we employed can be used by the community going forward as a means of fully understanding multi- device use alongside television watching.

Audience Silhouettes: Peripheral Awareness of Synchronous Audience Kinesics for Social Television

Radu-Daniel Vatavu – University Stefan cel Mare of Suceava, Suceava, Romania

Abstract: We introduce TV audience silhouettes, which are visual representations of viewers’ body movements and gestures displayed in real-time on top of broadcasted television. By using minimal visual cues, audience silhouettes are strong candidates for implementing Oehlberg et al.’s theater metaphor of an unobtrusive social TV system [34] by conveying presence and leveraging interaction via non-verbal kinesics. We found our study participants connecting well to the on-screen silhouettes, while their TV watching experience was perceived more enjoyable. We also report viewers’ body movement behavior in the presence of on-screen silhouettes, which we characterize numerically with average body movement, percentage of distinct body postures, and amplitude of movement, e.g., we found that the number of silhouettes influences viewers’ body movements and the body postures they adopt, with women producing more body movement than men.

Broadcast, Video-on-Demand, and Other Ways to Watch Television Content: a Household Perspective

Jeroen Vanattenhoven – CUO Social Spaces, iMinds – KU Leuven, Leuven, Belgium
David Geerts – CUO , Social Spaces, KU Leuven, iMinds, Leuven, Belgium

Abstract: This paper presents an investigation into they ways households currently use traditional broadcast television and video-on-demand services at the same time. The results were obtained via in-home interviews with seven households in The Netherlands. Both ways of watching television and video content still have their uses. Sometimes broadcast television is turned on out of necessity; at other times it provides a more valuable function than video-on-demand services. In this paper we go deeper into this matter and explain how each way of viewing TV content relates to different viewing situations in the home. These insights help us to formulate implications for the design of television products and services in general, and for better, contextual recommender systems.

Dynamic Subtitles: the User Experience

Andy Brown – BBC Research and Development, BBC, Manchester, United Kingdom
Rhianne Jones – BBC Research and Development, BBC, Manchester, United Kingdom
Michael Crabb – School of Computing, University of Dundee, Dundee, Scotland, United Kingdom
James Sandford – BBC Research and Development, BBC, Manchester, United Kingdom
Matthew Brooks – BBC Research & Development, BBC, Manchester, Greater Manchester, United Kingdom
Caroline Jay – School of Computer Science, University of Manchester, Manchester, United Kingdom
Mike Armstrong – BBC Research and Development, BBC, Manchester, United Kingdom

Abstract: Subtitles on television are typically placed at the bottom-centre of the screen. However, placing subtitles in varying positions, according to the underlying video content (`dynamic subtitles’), has the potential to make the overall viewing experience less disjointed and more immersive. This paper describes an investigation into the User Experience of dynamic subtitles. Qualitative data from habitual subtitle users demonstrates that dynamic subtitles can lead to an improved experience, although not for all types of user. Eye-tracking data was analysed to compare the gaze patterns of subtitle users with a baseline of those for people viewing without subtitles.

EnvDASH – An Environment-Aware Dynamic Adaptive Streaming over HTTP System

Stefan Wilk – Distributed Multimedia Systems, TU Darmstadt, Darmstadt, Hessen, Germany
Denny Stohr – Distributed Multimedia Systems, TU Darmstadt, Darmstadt, Hessen, Germany
Wolfgang Effelsberg – Distributed Multimedia Systems, TU Darmstadt, Darmstadt, Hessen, Germany

Abstract: Dynamic Adaptive Streaming over HTTP (DASH) is an approach to address changing network conditions during video streaming sessions. Our system differs from the standard as it extends DASH with mechanisms that allow sensing and adapting to the environment. By this our system introduce benefits regarding network load reduction in comparison to standard adaptations rules. The system senses whether the user is interested in watching a video, if the displaying device is held stable and the ambient noise. Depending on readings sensed our EnvDASH system allows to reduce produced data traffic. This is important for mobile users having volume contracts for cellular networks. Our work illustrates that unstable viewing conditions are common in mobile scenarios and that multi-modal analysis of the environment is able to reduce the data traffic in cellular networks. Additionally, as needs of users are very different users can easily customize rules while streaming video.

Experiencing Liveness of a Cherished Place in the Home

Jinyi Wang – Mobile Life @ Stockholm University, Kista, Sweden
Mudassar Ahmad Mughal – Mobile Life @ Stockholm University, Kista, Sweden
Oskar Juhlin – Mobile Life @ Stockholm University, Kista, Sweden

Abstract: Liveness, as discussed in HCI and in media studies, focuses on an intriguing and beloved experiential quality that can influence new forms of video applications. We suggest a shift from accounts of liveness in “events” to liveness in ambient media in home decor, by designing a system called TransLive that exploits the “magic” of mediatizing the “now” at a distant and cherished place. We present a field study including four families, who experienced the system for two weeks each in a living lab setting. It shows how immediacy and unpredictability provide compelling experiences. Authenticity and engagement, which are previously considered as inherent qualities in live media, instead occur in the context of use. Finally, the experience of transcendence triggered by slow and continuous video streams and sensor data, unveils a new design space of liveness. Thus, we not only need to take inspiration from liveness theory, but also redefine it.

Experimental Enquiry into Automatically Orchestrated Live Video Communication in Social Settings

Marian Ursu – Department of Theatre, Film and Television, University of York, York, United Kingdom
Manolis Falelakis – Goldsmiths, University of London, London, United Kingdom
Martin Groen – Department of Languages, Literature and Communication, University of Utrecht, Utrecht, Netherlands
Rene Kaiser – Institute for Information and Communication Technologies, JOANNEUM RESEARCH, Graz, Austria
Michael Frantzis – Goldsmiths, University of London, London, United Kingdom

Abstract: This paper asserts that more complex setups of live video mediated social communication and interaction should be able to dynamically adapt to the specific communication contexts they mediate. This feature is referred to as ‘orchestration’. Previous research started to explore this concept and, by carrying out orchestration through human operators, evidence was found indicating that it could improve the quality of the communication experience. The study reported here continues that line of enquiry, but employing also a completely automatic system. A similar conclusion was inferred here, too, namely that automatic orchestration can improve the quality of the communication experience. The evidenca was based on an objective measure of ‘task efficiency’, defined on the basis of the points scored in the social game that framed the experience. However, two other measures employed, a questionnaire (subjective) and average number of turnshifts and turn duration (objective), were inconclusive. At this end, the paper also uncovers some of the complexities of the conceptual space associated with orchestrated mediated communication. Finally, the paper also aims to provide motivation for further research into this communication paradigm.

First Person Omnidirectional Video : System Design and Implications for Immersive Experience

Shunichi Kasahara – SonyCSL, Shinagawa, Tokyo, Japan
Shohei Nagai – The University of Tokyo, Toyko, Japan
Jun Rekimoto – University of Tokyo / Sony CSL, Bunkyo-ku, Tokyo, Japan

Abstract: Fully recording and sharing an immersive experience is one of the ultimate goals of media technology. As extensive technical evolution, omnidirectional video is one of promising media to capture an immersive experience. First person omnidirectional video provide unique experience of world through someone else’s perspective. This will bring various applications such as entertainment, sports viewing, education and simulation training, newscasting and therapy. However, difficulties in wearable camera design and cybersickness induced by shaky video has been obstacle to explore applications of first person omnidirectional video. In this research, we introduce the design and implementation of “LiveSphere” a system including a wearable omnidirectional camera and image stabilization to improve cybersickness. Our evaluation revealed the improvement of cybersickness. Then we report the series of workshops to explore user experience and applications in actual use cases such as virtual travel and virtual sports. we have compiled design implications about cybersickness and motion, immersive sensation, visualization and behavior data of spectators in experience with first person omnidirectional video.

Interactive UHDTV at the Commonwealth Games – An Explorative Evaluation

Judith A. Redi – Delft University of Technology, Delft, Netherlands
Lucia D’Acunto – TNO, Delft, Netherlands
Omar Niamut – Media & Network Services, TNO, Delft, The Netherlands

Abstract: In conjunction with BBC R&D experiments and demonstrations at the 2014 Commonwealth Games, an explorative field trial was conducted with a live zoomable UHD video system. The unique field trial featured the world’s first live tiled streaming of 4K UHD video to end users. During the trial, we studied and evaluated the attractiveness and novelty of an interactive UHD application, and investigated system design aspects of a live UHD tiling system. In this paper, we evaluate the overall perceived quality of experience (QoE) of the application and to what extent the QoE depends on system factors and/or network conditions. We observe that interactive UHDTV is well received by users, but the delivered experience may decrease in presence of low bandwidth availability.

It Takes Two (To Co-View): Collaborative Multi-View TV

Mark McGill – School of Computing Science, University of Glasgow, Glasgow, United Kingdom
John Williamson – School of Computing Science, University of Glasgow, Glasgow, United Kingdom
Stephen Brewster – School of Computing Science, University of Glasgow, Glasgow, United Kingdom

Abstract: This paper investigates how we can design interfaces and interactions for multi-view TVs, enabling users to transition between independent and shared activity, dynamically control awareness of other users’ activities, and collaborate more effectively on shared activities. We conducted two user studies, first comparing an Android-based two-user TV against both multi-screen and multi-view TVs. Based on our findings, we iterated on our design, giving users the ability to transition between casual and focused modes of usage, and dynamically set their engagement with other users’ activity. We provide the foundations of a multi-user multi-view smart TV that can support users to transition between independent and shared activity and gain awareness of the activities of others, on a single shared TV that no longer suffers the bottleneck of one physical view, significantly improving upon a user’s capability for collaborative and independent activity compared to single-view smart TVs.

Towards an Extended Festival Viewing Experience

Raphael Velt – Mixed Reality Lab, The University of Nottingham, Nottingham, United Kingdom
Steve Benford – Mixed Reality Lab, University of Nottingham, Nottingham, United Kingdom
Stuart Reeves – School of Computer Science, The University of Nottingham, Nottingham, United Kingdom
Michael Evans – BBC Research & Development, British Broadcasting Corporation, Manchester, United Kingdom
Maxine Glancy – Research & Development, BBC, Salford, Greater Manchester, United Kingdom
Phil Stenton – BBC Research & Development, BBC, Salford, United Kingdom

Abstract: Media coverage of large-scale live events is becoming increasingly complex, with technologies enabling the delivery of a broader range of content as well as complex viewing patterns across devices and services. This paper presents a study aimed at understanding the experience of people who have followed the broadcast coverage of a music festival. Our findings show that the experience takes a diversity of forms and bears a complex relationship with the actual experience of being at the festival. We conclude this analysis by proposing that novel services for coverage of this type of events should connect and interleave the diverse threads of experiences around large-scale live events and consider involving more diverse elements of the experience of ‘being there’.

Who’s The Fairest Of Them All: Device Mirroring For The Connected Home

Mark McGill – School of Computing Science, University of Glasgow, Glasgow, United Kingdom
John Williamson – School of Computing Science, University of Glasgow, Glasgow, United Kingdom
Stephen Brewster – School of Computing Science, University of Glasgow, Glasgow, United Kingdom

Abstract: In the UK alone smartphone adoption has reached 61\% in 2014. In home and living-room contexts, this adoption has led to “multi-screening”, meaning the concurrent use of devices such as smartphones and tablets alongside the TV. The resultant private “digital bubble” of this device usage has been discussed as raising a problematic barrier to socialization and interaction, with mobile phone use in particular having significant anti-social connotations. However mobile devices have evolved new capabilities for sharing their activity, most notably through screen mirroring. This paper explores how we can utilize the TV to view screen-mirrored device activity, decreasing the digital isolation of device usage. We examine the extent to which users can attend to multiple devices on one TV, the effect this and prior systems have had on existing TV viewing, and propose ways in which we can aid users to manage their viewing of device activity on the TV. Moreover, we examine new approaches toward the accessibility of device activity, investigating systems which allow users to attend to whichever device activity they wish using multi-view displays, and discuss the social and privacy implications of having “always-on” screen-mirrored devices.