Appeal 2007-1658
Application 09/838,428
is whether the combination of Loui and Bergen teaches or suggests
extracting a first video frame associated with a first set of text data and a
second video frame associated with a second set of text data and forming
two still images therefrom.
Loui discloses (col. 1, ll. 61-67) that video clips can be placed in a
digital album by selecting a key frame for static display to identify the video.
Thus, Loui suggests extracting (by selecting) a video frame (a key frame)
from plural video frames (the video clip) to display a frame representative of
the video clip. Further, Loui discloses (col. 2, ll. 1-5) that modern cameras
allow associating textual data with digital images. Loui (col. 5, ll. 40-48)
describes display 20 of Figure 3 as showing four digital photographs, each
with associated text describing the photograph. Thus, Loui suggests
associating text with the images to describe them.
Bergen (col. 2, ll. 29-32 and 44-47) describes a database that provides
scene-based video information to a user by dividing a video stream into
scenes, each made up of plural frames, with a key frame for each scene.
Bergen further discloses (col. 4, ll. 12-21) providing information associated
with the video to identify portions of one or more frames or scenes. The
information may be summaries or textual descriptions of the scenes. (See
col. 10, ll. 31-36.) Thus, Bergen further suggests extracting a key frame
from each of a plurality of video streams and associating text with each
extracted frame. We note that Appellants argue (Br. 17) that Bergen "does
not teach or suggest dividing a video stream based on sets of text data
associated in time with moving video frames." However, independent
claims 1, 8, and 15 do not require dividing the video stream "based on" text
data. The claims merely call for extracting two video frames associated with
4
Page: Previous 1 2 3 4 5 6 7 Next
Last modified: September 9, 2013