Mapping GOPS in an Improved DVC to H.264 Video Transcoder - Advanced Concepts for Intelligent Vision Systems - page 26

This idea can be extended for longer WZ GOPs but some considerations must be
taken into account. Figure 5 shows the transcoding process for a WZ GOP 4. As it is
shown in the two top rows, WZ decoding algorithm divides the decoding process into
two steps. In the first step, WZ 2 is decoded by calculating the SI between K 0 and K 4 .
These MVs (V 0-4 ) are ignored as they have low accuracy. In the second step, there is a
reconstruction of WZ 2 (labeled WZ' 2 ) and now this case is similar to the previous one
showed in Figure 4. Then, V 0-2 and V 2-4 are divided to improve H.264 encoding
procedure. This procedure can be applied for any WZ GOP, including odd GOPs.
Fig. 5. Mapping from DVC GOP 4 to H.264 GOP I11P
3.2 Motion Estimation Reduction
DVC to H.264 transcoding process joins the largest complexity algorithms of each
paradigm, so a lot of effort must be invested in order to improve this task. As it is well
known, a big part of this complexity depends largely on the search range used in the
H.264 ME process is as a consequence of the quantity of checking done. However,
this process may be accelerated because of the search range can be reduced avoiding
unnecessary checking without significant impact on quality and bit rate.
To achieve this aim, we propose to reuse the MVs calculated in DVC to define a
smaller search range for each MB of H.264 including every sub MB partition. So in
this way the checking area is limited by the area S defined in the expression (1).
(1)
where (x,y) are the coordinates to check, A is the search range used by H.264 and C is
a circumference which restricts the search with centre on the upper left corner of the
MB. C is defined by the equation (2).
(2)