Appeal No. 2006-1618
Application No. 10/046,797
Appellant further contends that no reasonable expectation of success
exists in combining Kim with Suzuki since slowdowns in Kim's system caused by
user intervention would destroy or frustrate Kim's purpose (i.e., encoding entire
frames of video) by encoding user-selected objects. Such a modification of Kim,
according to appellant, would result in unacceptable encoding speeds [brief,
page 10; reply brief page 4].
The examiner responds that Kim and Suzuki are properly combinable
since both references detect image contours -- albeit for different purposes
[answer, page 4]. The examiner further notes that applying the teachings of
Suzuki would not slow Kim's system unacceptably since Suzuki does not require
user input in every frame of video; rather, only one user input is required. After
such user input, contours are then detected automatically in subsequent frames
[answer, pages 4 and 5].
The examiner further notes that the motivation to combine the references
was expressly stated in Suzuki, namely so that the user can extract a single
object from a sequence of images of a motion picture on the basis of a detected
object contour. According to the examiner, without such a user input, Kim's video
encoder would be unable to select a single object for extraction [answer, page 5].
We will sustain the examiner's rejection of claim 1. We note at the outset
that Kim discloses selecting two starting vertices (A and B) depending on
whether the image is an open or closed loop [Kim, col. 3, lines 13-18; Fig. 2A].
Kim does not state that the starting vertices A and B are manually selected, but
7
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Next
Last modified: November 3, 2007