45
C H A P T E R 5
CCFM Decoding and
Visualization
e CCFMs as introduced in the previous chapter are not just a powerful feature representation
generalizing other popular descriptors, but also allow for the extraction of robust estimates. For
instance, the pose vector estimates described previously or the results from channel smoothing
(see Figure 5.1) are obtained by the decoding of CCFMs, which is the central topic of this chapter.
x
Smoothed
Signal
Decode
xx
Signal
Channels
x
Averaged Channels
Local Averaging of Channels
Metamery
Region
Encode
Figure 5.1: Illustration of channel smoothing: the noisy signal (left) is smoothed without blur-
ring the edge (right). is is achieved by spatial averaging of the channels, leading to a metamery
region. Illustration by Felsberg and Granlund [2003].
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset