Page 29 - Kaleidoscope Academic Conference Proceedings 2022
P. 29
the aforementioned Recommendation ITU-T P.1320, new by generating a shared virtual space that is geometrically
subjective evaluation methodologies must be added, as consistent with the physical space in which the different
well as objective metrics for monitoring and dimensioning participants move.
the network. ITU-T has already started addressing this Subsequently, we have proposed and analyzed a network
problem, with the analysis of QoE influencing factors for architecture to deploy distributed reality services in a 5G and
VR (G.1035 [48]) and AR (G.1036 [49]), as well as with B5G network. We have described the basic quality models
a methodology to assess the quality of immersive video that exist in the state of the art, and we have proposed a
(P.919 [50]). However, the subjective evaluation of immersive way to be able to have coarse parametric models that allow
communications systems is a significantly more complex task improving the dimensioning of the network.
than simply assessing the visual quality offered by the HMD. Finally, we have analyzed the main opportunities for
First, the ability of XR technology to elicit a sense of standardization that arise from this technology: on the
presence, even with much poorer visual quality than that one hand, in the design of an interoperable system (using
available in, say, high-definition video on a screen, makes telephony as an example) and, on the other, in the
it not enough to measure subjective quality-of-the-pixel. development of subjective methodologies and objective
It is also necessary to measure socio-emotional aspects metrics to assess the quality of experience.
such as the feeling of presence or the ability to improve
empathy in communications [17]. In order to carry out ACKNOWLEDGEMENT
experiments that allow measuring differences in these aspects,
more sophisticated methodologies are needed, which include The author would like to acknowledge the whole Nokia
interaction between several people and carrying out different eXtended Realty Lab, as well and the Image Processing
communication tasks [11]. The challenge ahead is being Group (GTI) at Universidad Politécnica de Madrid, since this
able to develop a Recommendation that is precise enough keynote paper shows the work done jointly with researchers
to define a set of repeatable tasks and measurements across of both teams. Special thanks to Marta Orduna, Carlos
multiple labs, while also being broad and complex enough to Cortés, Diego González-Morín, and Ester González-Sosa,
cover many types of systems and measure various perceptual from whom I have learnt so much in the last few years.
and socioemotional properties. The Immersive Media This work has received funding from the Spanish
Group (IMG) of the Video Quality Experts Group (VQEG) Administration Agency Centro para el Desarrollo
is currently starting a project to design and validate an Tecnológico Industrial (CDTI) under project IDI-20210977
assessment methodology that can address this challenge for a (IMMINENCE) as an International Technical Cooperation
wide range of immersive communication systems. R&D Project.
Finally, with regard to objective quality models, the range
of pending work is wide. Today there are no sufficiently REFERENCES
consolidated metrics that allow us to accurately model, [1] J. G. Apostolopoulos, P. A. Chou, B. Culbertson,
measure or monitor the objective quality of an immersive T. Kalker, M. D. Trott, and S. Wee, “The road to
communication system. No doubt new Recommendations immersive communication,” Proceedings of the IEEE,
will emerge over the years that address this problem. In the vol. 100, no. 4, pp. 974–990, 2012.
meantime, there is only recourse to simple solutions like the
one outlined in Section 3.2. [2] I. Rae, G. Venolia, J. C. Tang, and D. Molnar,
“A framework for understanding and designing
telepresence,” in Proceedings of the 18th ACM
5. CONCLUSIONS
conference on computer supported cooperative work &
Throughout this paper we have explored the idea of the social computing, pp. 1552–1566, 2015.
realverse and its implications. First of all, we have proposed [3] J. McVeigh-Schultz, A. Kolesnichenko, and K. Isbister,
the realverse as the vision that several people in distant places “Shaping pro-social interaction in VR: an emerging
can meet each other and interact as if they were visiting the design framework,” in Proceedings of the 2019 CHI
same physical place. The concept of the realverse is closely Conference on Human Factors in Computing Systems,
linked to its practical implementation, distributed reality: pp. 1–12, 2019.
combining audiovisual content captured in different places,
in such a way that it gives rise to a consistent immersive [4] B. Ens, J. Lanir, A. Tang, S. Bateman, G. Lee,
experience, allowing people to interact naturally. T. Piumsomboon, and M. Billinghurst, “Revisiting
Throughout the text we have also proposed a way to collaboration through mixed reality: The evolution of
implement a first version of the realverse from its fundamental groupware,” International Journal of Human-Computer
Studies, vol. 131, pp. 81–98, 2019.
building blocks. Our previous work on immersive
telepresence with 360 video and on video-based avatars shows [5] A. Villegas, P. Pérez, and E. González-Sosa, “Towards
that this technology provides a very high level of presence, a distributed reality: a multi-video approach to XR,” in
even though it still has technical limitations. We have also Proceedings of the 11th ACM Workshop on Immersive
seen that these components must be completed by including Mixed and Virtual Environment Systems, pp. 34–36,
video-based avatars also for remote users and, above all, 2019.
– xxv –