Image Processing Reference
In-Depth Information
Fig. 5.10 Display scalable 3D holoscopic video coding architecture using three hierarchical
layers
Another important requirement for allowing efficient storage and delivery of 3D
holoscopic content to the end-users is high compression efficiency. Regarding the
presented display scalable coding architecture, this would mean significant perfor-
mance gains compared to independent compression of each coding layer (the
simulcast case).
For this reason, the coding information flow in this 3D holoscopic coding
scheme (for one access unit, i.e., for all pictures that represent the video scene at
the same time instant) is defined as the following:
1. In the Base Layer , the 2D views are coded with a suitable standard 2D video
encoder and the reconstructed frames are kept for coding the upper layers.
2. Between the Base Layer and the First Enhancement Layer , an inter-layer
prediction mechanism exploits the existing inter-view correlation to improve
the coding efficiency. Similarly, within the First Enhancement Layer an inter-
view prediction is also used and the encoded and reconstructed data is fed to the
Second Enhancement Layer. It should be noticed that, efficient prediction
mechanisms between the Base Layer and the First Enhancement Layer and
within the First Enhancement Layer are not addressed in this chapter. However,
these cases have been extensively studied in the context of stereo and Multiview
Video Coding (MVC) [ 16 ] extensions of the H.264/AVC [ 17 ], and the proposed
3D video coding extensions of the High Efficiency Video Coding (HEVC) [ 18 ]
standard for multiview video and depth data [ 19 ]. For a good review of these 3D
video coding solutions, the reader can refer to [ 20 - 22 ].
3. Between the First and the Second Enhancement Layers , an inter-layer prediction
method, which is described in Sect. 5.4.1 , is used to exploit the optical-geometric
relation between the multiview content and the 3D holoscopic content. Never-
theless, it was shown in [ 23 ] that further improvements can still be reached by
integrating an efficient prediction scheme that explores the inherent spatial
redundancy of the 3D holoscopic content itself. For this reason, within the
Search WWH ::




Custom Search