Appeal No. 1999-2018
Application 08/571,204
Appellant argues that "unlike the present method of
claim 1 which selects a predictive coding mode based upon
the values of the first and second inter picture distances,
Sugiyama selects a prediction option based on the lowest
prediction error" (Br8).
We interpret the "inter picture distances" in claim 2
as broad enough to read on the forward and backward
estimation error values in Sugiyama which are used to select
the three predictive coding modes. Claim 2 does not define
"distance" as a count of the number of pictures from a past
or future reference picture. Therefore, while Appellant's
invention, as disclosed, is different from Sugiyama, claim 2
is so broad that it is anticipated by Sugiyama for this
additional reason.
Appellant's specification defines the forward
prediction error (motion vector estimation error between a
B-picture and a past reference picture) as Ef, and the
backward prediction error (motion vector estimation error
between a B-picture and a future reference picture) as Eb
(specification, p. 6), where Ef and Eb are a measure of the
correlation degree; the smaller the number, the greater the
- 10 -
Page: Previous 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Next
Last modified: November 3, 2007