Introduction - Novel 3D Media Technologies - page 3

considering the extent to which the number of such sub-media processing tasks can
reach in line with the applications as well as the user population, in such shared
environments, how to appropriately provision computing resources to different
tasks dynamically with regards to both system efficiency and Quality of Service
remains a challenging problem. Different processing tasks, such as core multimedia
encoding standards (e.g. H.264, HEVC, etc.) and computational tasks (e.g. Motion
Estimation/Motion Compensation, DCT calculation, etc.), packetisation,
transcoding, retargeting including transform domain and pixel domain processing,
rendering (e.g. free-viewpoint video rendering), adaptation and mash-up with other
content sources (e.g. web applications or other User Generated Content) have
different computational, memory and bandwidth intensities. Most of the existing
solutions focus only on maximising the system efficiency considering generic tasks,
without necessarily overseeing the QoS requirements of various media processing
sub-tasks and in overall the QoE offered to the end user. In that sense, the
deployment of dynamic resource allocation within cloud architectures has been
favoured to static resource allocation. Machine learning-based resource allocation
for media processing tasks aims at getting a more precise prediction for various
multimedia processing related sub-tasks based on historical data (e.g. average
memory footprint, variance over time for various applications contexts, etc.).
This topic describes recent innovations in 3D media and technologies, with
coverage of 3D media capturing, processing, encoding, and adaptation, networking
aspects for 3D Media, and quality of user experience (QoE). The structure of the
book is the following:
Chapter 2 is entitled “Novel approaches to Immersive Media: from enlarged
field-of-view to multi-sensorial experiences”. The chapter presents a review of
current evidence on the influence of immersion (defined in terms of the technical
features of the system) on the user experience in multimedia applications. The
chapter introduces the concepts of media enjoyment, presence and Quality of
Experience (QoE) that frame the analysis from the user perspective. It discusses
the bounding effects of multimodal perception on the previously defined metrics.
Furthermore, it analyses the influence of relevant technical factors on presence,
enjoyment and QoE, with emphasis on those characterising the level of immersion
delivered by system across four dimensions: inclusiveness, extensiveness, sur-
rounding and vividness. Finally, it presents recent works integrating some of
these factors into multi-sensorial media experiences and highlights open issues
and research challenges to be tackled in order to deliver cost-effective multi-
sensorial media solutions to the mass market.
Chapter 3 is entitled “3D video representation and coding”. The technologies
which allow an immersive user experience in 3D environments are rapidly evolving
and new services have emerged. Most of these services require the use of 3D video,
combined with appropriate display systems. As a consequence, research and devel-
opment in 3D video continues attracting sustained interest. While stereoscopic
viewing is already widely spread, namely in TV and gaming, new displays and
applications, such as FTV (Free viewpoint TV), require the use of a larger number
of views. Hence, the multi-view video format was considered, which uses N views,