On the characteristics of the sequence generated by a sliding window on MPEG-4 FGS video and its use in streaming

In several applications of video, a sliding window is used to generate a new sequence for further processing. If the width of the sliding window is fixed and equal to that of a Group of Pictures(GoP), the new sequence generated by the sliding window is seen to have some interesting properties. In particular, it is seen that the samples of the new sequence come from a single normal distribution, while the samples of the original video sequence come from different distributions with different means and variances. This results in less variations between adjacent samples for the new sequence. The usefulness of this reduced variability is illustrated through a video streaming application. The streaming algorithm looks at the future bit requirement of the video by maintaining a sliding window, and it is seen that the reduced variability of the sequence helps in making more accurate capacity planning.