Image Processing Reference
In-Depth Information
Second Enhancement Layer
, a self-similarity compensated prediction (described
in
4.2
) is integrated so as to take also full advantage of the inherent significant
cross-correlation between neighboring micro-images.
Since 3D holoscopic video can be interpreted as a sequence of 2D frames
(arranged as arrays of micro-images), a simple coding approach by using a regular
2D video encoder could be used. For this reason and since the presented scalable
codec has most of the design elements in common with the hybrid coding tech-
niques of HEVC [
18
], a 3D holoscopic scalable coding extension is easily enabled
for this 2D video coding standard as explained in more details in Sect.
5.4.3
.
5.4.1
Inter-Layer Prediction Scheme
It was shown in [
15
] that higher coding efficiency can be achieved by exploring the
existing redundancy between the
First Enhancement Layer
and the
Second Enhance-
ment Layer
, through an inter-layer prediction scheme. This prediction scheme builds
an inter-layer (IL) reference picture which is then used to predict a 3D holoscopic
image being coded. To build an IL reference, the following data are needed:
•
Set of 2D views
: The set of reconstructed 2D views from the previous coding
layers are obtained by decoding the bitstream generated for the lower layers;
•
Acquisition parameters
: These parameters comprise information from the 3D
holoscopic capturing process (such as the resolution of micro-images and the
structure of the micro-lens array) and also information from the 2D view
generation process (i.e., size and position of the patch, as explained in
Sect.
5.3.4
). This information has to be conveyed along with the bitstream to
be available at the decoding side.
Therefore, two steps are distinguished when generating an IL reference:
Patch
Remapping
, and
Micro-Image Refilling
.
5.4.1.1 Patch Remapping Step
This first step is an inverse process of the Basic Rendering algorithm, presented in
Sect.
5.3.4
, i.e., it corresponds to an inverse mapping (referred to as remapping)
of the patches from all rendered and reconstructed 2D views to the 3D holoscopic
image.
For this step, the needed input information is:
- One or more reconstructed 2D views;
- Patch size (PS) used to generate the 2D views;
- Micro-image resolution;
- Relative position of the patch (PP) in the micro-image used to take each different
2D view.
