Multiview 3D videos can be represented explicitly or implicitly. In an explicit representation, all possible
views are either coded separately (simulcast coding) or jointly using multiview coding [Vetro et al.
2011]. Using only texture information to drive multiview displays requires providing a large number
of views to the display. This is not efficient since it requires transmitting a large amount of data which
can exceed the network capacity. Even if multiview coding is used to encode the views, the resulting
bit rate will still be proportional to the number of views [Muller et al. 2011]. Moreover, it has been ¨
shown that for common multiview coding conditions only up to 30% of the macroblocks are predicted
from inter-view reference pictures