Appeal No. 2004-1628
Application 10/254,720
and at column 5, lines 8-28, and column 8, line 49 through column 9, line 2, wherein a “low and
high texture type/pattern areas are assigned/identified by variance processor 8 (answer-page 4).
We have reviewed the Singhal reference and find that while the reference does show a
variance which is indicative of the texture in an image, it does not suggest “assigning a texture
type to each texture area having a similar texture pattern,” as claimed. Rather, as argued by
appellant, at page 5 of the brief, Singhal “only discloses how different textures effect coding.”
This is borne out by Singhal, at column 5, lines 23-27, wherein it is discussed that bits are
allocated for coding based, in part, on the texture of the macroblocks. Column 5, lines 16-18, of
Singhal, also points out that variance “serves as a good indicator of the activity or texture in the
image” but it does not teach or suggest that a texture type is “assigned” to each texture area
having a similar texture pattern.
Column 8, lines 49 et seq. of Singhal discusses low and high texture patterns in a
macroblock and the application of small and large quantizing step sizes, but, again, we find
nothing therein indicative of the claimed “assigning a texture type to each texture area having a
similar texture pattern.”
4
Page: Previous 1 2 3 4 5 6 7 8 9 Next
Last modified: November 3, 2007