Information Technology Reference
In-Depth Information
(a)
(b)
(c)
(d)
×
Fig. 3
Frame doubling of the 130
100 sequence
Cameraman Pan
.
a
original frame 5 (9
in the output). New frame 10 by
b
frame averaging,
c
variational TSR without GCA, and
d
variational TSR with GCA.
uniform global translational motion due to the depth of the scene and variations in
the motion of the camera. Frame averaging on
Building
blurs the new frames
quite a lot as can be seen in Fig. 4(b). During video playback [32] this blurring
is not sensed, but the motion is as jerky as in the input. The lamps seen inside
the building seems to flicker as they are blurred to middle grey in the new frames.
These artifacts make the frame averaging result very annoying to watch. The motion
portrayal in the two variational TSR results is natural and no flickering occurs. As
seen in Figs. 4(c) and 4(d) the new frames are a bit smoothed when compared to
