Ex Parte Zhang

Patent Trial and Appeal Board

Feb 26, 2013

10807949 (P.T.A.B. Feb. 26, 2013)

UNITED STATES PATENT AND TRADEMARK OFFICE
_____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
_____________

Ex parte TONG ZHANG
_____________

Appeal 2010-007555
Application 10/807,949
Technology Center 2400
______________

Before JEAN R. HOMERE, BRYAN F. MOORE, and JOHN G. NEW,
Administrative Patent Judges.

MOORE, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-005291
Application 10/489,076

2
This is a decision on appeal under 35 U.S.C. § 134(a) of the Final
Rejection of claims 23-48 which represent all the pending claims. App. Br.
1. We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.

INVENTION
The invention is directed to a method for intelligent extraction of key-
frames from a video that yields key-frames that depict meaningful content in
the video. See Spec. 3:3-5.
Claim 23 is exemplary of the invention and is reproduced below:
23 A key-frame extraction system, comprising:
video frame extractor that extracts each of a series of
video frames from a video;
a set of frame analyzers that obtain the series of video
frames in parallel from the video frame extractor, each frame
analyzer selecting a corresponding set of candidate key-frames
from the series by performing a different corresponding
analysis on each video frame in the series such that the analyses
are selected to 10 detect multiple types of meaningful content in
the video;
key-frame selector that obtains the corresponding
candidate key-frames from each frame analyzer and arranges
the candidate key-frames into a set of clusters and that selects
one of the candidate key-frames from each cluster as a key-
frame for the video.

REFERENCES
Dimitrova US 6,125,229 Sep. 26, 2000
Wu US 2003/0068087 A1 Apr. 10, 2003
Toklu US 6,549,643 B1 Apr. 15, 2003

Appeal 2010-005291
Application 10/489,076

3
REJECTIONS AT ISSUE
Claims 23, 25, 26, 28-33, 36, 38, 39, and 41-46 stand rejected under
35 U.S.C. § 102(e) as anticipated by Toklu. Ans. 4-6.
Claims 24, 27, 34, 37, 40, and 47 stand rejected under 35 U.S.C.
§ 103(a) as unpatentable over the combination of Toklu and Wu. Ans. 7-9.
Claims 35 and 48 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over the combination of Toklu and Dimitrova. Ans. 9-10.

ISSUES
1. Did the Examiner err in finding that Toklu discloses “a key-
frame selector that obtains ... and arranges the candidate key-
frames into a set of clusters and that selects one of the candidate
key-frames from each cluster as a key-frame for the video” as
recited in claim 23; and
2. Did the Examiner err in finding that Toklu discloses “set of
frame analyzers that obtain the series of video frames in parallel
from the video frame extractor [and] performing a different
corresponding analysis on each video frame in the series . . .” as
recited in claim 23; and
3. Did the Examiner err in finding that Toklu discloses “key-frame
selector selects the key-frames by determining an importance
score for each candidate key-frame in each cluster” as recited in
claim 25; and
Appeal 2010-005291
Application 10/489,076

4
4. Did the Examiner err in finding that Toklu discloses “the key-
frame selector determines the importance scores by determining
an image content of each candidate key-frame” as recited in
claim 26; and
5. Did the Examiner err in finding that Toklu discloses “the key-
frame selector selects the key-frames by determining an image
quality for each candidate key-frame in each cluster” as recited
in claim 28; and
6. Did the Examiner err in finding that Toklu discloses “the frame
analyzers include a color layout analyzer” as recited in claim
30; and
7. Did the Examiner err in finding that Toklu discloses “the frame
analyzers include a fast camera motion detector” as recited in
claim 31; and
8. Did the Examiner err in finding that Toklu discloses “wherein
the frame analyzers include an object motion analyzer” as
recited in claim 33; and
9. Did the Examiner err in not explicitly describing the level of
ordinary skill in the art regarding the combination of Toklu and
Wu; and
10. Did the Examiner err in finding one of ordinary skill in the art
would be motivated to combine Toklu and Wu; and
Appeal 2010-005291
Application 10/489,076

5
11. Did the Examiner err in not explicitly describing the level of
ordinary skill in the art regarding the combination of Toklu and
Wu; and
12. Did the Examiner err in finding one of ordinary skill in the art
would be motivated to combine Toklu and Wu; and
13. Did the Examiner err in finding the combination of Toklu and
Dimitrova teaches “previous to each key-frame” and “selection
of one or more of the video frames” as recited in claim 35?

ANALYSIS
35 U.S.C. § 102(e) – Toklu
Claims 23, 25, 26, 28-33, 36, 38, 39, and 41-46
Claim 23
Appellant argues Toklu fails to disclose “a key-frame selector ‘that
obtains ... and arranges the candidate key-frames into a set of clusters,. . .’ as
recited in claim 23.” App. Br. 21. Specifically, Appellants argue that the “the
segments according to Toklu et al. are already arranged prior to any attempt
to extract key-frames.” Id.at 22. This argument is not persuasive.
Toklu discloses a Video Segmenter 12 which arranges the video in
segments which “represent a contiguous recording of one or more video
frames depicting a continuous action in time and space.” Toklu 5:55-57. The
video is further processed until a Key-Frame Elimination module 18
eliminates similar key frames and then stores them based on the segment
arrangement established by the Video Segmenter 12. See, Toklu, 7:63-65
(“After the key-frame elimination process, the remaining key-frames for
Appeal 2010-005291
Application 10/489,076

6
each segment are output in a key-frame list (step 210).”). Therefore, we find
ample support for the Examiner’s finding that Toklu discloses “a key-frame
selector that obtains ... and arranges the candidate key-frames into a set of
clusters.”
Claim 23 also recites a key frame selector “that selects one of the
candidate key-frames from each cluster as a key-frame for the video.”
Appellant argues that this requires the key frame selector to select one and
only one key-frame for the video. We are not persuaded by this argument. At
the outset, we note that the preamble of claim 23 recites the term
“comprising,” which is an open-ended transition phrase. It has been held that
when used in claim drafting, the term “comprising” indicates that the claim
is open-ended such that while named features in the claim are essential other
unrecited features are not excluded. See Gillette Co. v. Energizer Holdings,
Inc., 405 F.3d 1367, 1371-1372 (Fed. Cir. 2005); see also Genentech, Inc. v.
Chiron Corp., 112 F.3d 495, 501 (Fed. Cir. 1997). Furthermore, the
indefinite article “a” when used to introduce a feature in an open-ended
claim usually carries the meaning “one or more.” KCJ Corp. v. Kinetic
Concepts, Inc., 223 F.3d 1351, 1356 (Fed. Cir. 2000). It is only in rare
circumstances that the article “a” is limited to a singular interpretation and
such limitation must be clearly established. Id. Such a circumstance may
arise when the specification or the claim language itself indicates that “a”
does not carry a meaning of one or more but instead means “one and only
one.” Harari v. Lee, 656 F.3d 1331, 1341 (Fed Cir. 2011). We thus consider
the meaning of “a” against the backdrop of the claim language and involved
specification, keeping in mind that, during prosecution, claims must be given
their broadest reasonable interpretation consistent with the specification.
Appeal 2010-005291
Application 10/489,076

7
We construe this limitation to require that “a,” i.e., at least one, key-
frame is selected, as recited in the claim language, rather than one and only
one key-frame. We find the Examiner has shown that in Toklu the key-frame
selector 18 selects one of the candidate key-frames, i.e., a reference key-
frame, from each segments as “a,” i.e., at least one, key-frame for the video.
See Toklu, Fig. 4; 13:60-66.
Appellant argues that “Toklu et al. do [sic] not disclose ‘a set of frame
analyzers’ that operate in parallel to perform ‘different corresponding
analysis’ where each frame analyzer selects ‘a corresponding set of
candidate key-frames,’ as recited in Claim 23.” App. Br. 7. We are not
persuaded by this argument. Toklu discloses frame analyzers 15, 16, and 17,
which each perform different analysis such as “Motion Detection Analysis”
or “Color Histogram Computation and Analysis.” Toklu, Fig. 1; see also
Ans. 12. Toklu further discloses that “these processes may be performed one
at a time (in any order) or in parallel.” Toklu, 12:49-50; see also Ans. 12.
Thus, we find ample support for the Examiner’s finding that Toklu discloses
“set of frame analyzers that obtain the series of video frames in parallel from
the video frame extractor [and] performing a different corresponding
analysis on each video frame in the series.”
Claim 36 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 36 for the reasons
stated above.
Claim 25
Claim 25 recites that the “key-frame selector selects the key-frames
by determining an importance score for each candidate key-frame in each
Appeal 2010-005291
Application 10/489,076

8
cluster.” The Examiner relies on a determination of object motion to show
the importance score. Ans. 13. Appellant argues that importance score is
defined in the Specification as a score based on a set of characteristics of the
candidate key-frame that may include among a number of different
characteristics whether the candidate key-frame satisfies one of the camera
motion rules of the camera motion tracker. App. Br. 11 (citing Spec. 21:25-
23:12). Based on that definition, Appellant argues that “the term, ‘object
motion’ as employed by Toklu et al. is not equivalent to an importance
score.” Id.
Toklu, as part of key-frame elimination, determines whether
significant motion has occurred in order to select or eliminate key-frames.
Toklu 13:57-67 (“If the object motion is determined to be insignificant
(negative result in step 405), then the comparison key-frame is eliminated
(step 406) . . . On the other hand, if the object motion is determined to be
significant (affirmative decision in step 405), then the reference key-frame is
eliminated.”); see also Ans. 13. In other words, even given Appellant’s
definition of an importance score, Toklu uses a characteristic of the key-
frame, i.e., the significance of motion in the frame, to score certain frames as
either passing, therefore included, or failing, therefore eliminated. Therefore,
we are not persuaded by this argument.
Claim 38 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 38 for the reasons
stated above.
Appeal 2010-005291
Application 10/489,076

9
Claim 26
Claim 26 recites that “the key-frame selector determines the importance
scores by determining an image content of each candidate key-frame.”
Appellant argues that although Toklu discloses “segmenting ‘the image into
regions of similar color (step 401)[,’ s]egmenting the image into regions of
similar color is not equivalent to ‘determining an image content’ which
generally includes much more than just regions of similar color.” App. Br.
12. Appellant does not explain the reason for defining image content as
much more than just regions of similar color, therefore, we are not
persuaded by this argument. See Ex parte Belinne, No. 2009-004693, 2009
WL 2477843 at *3-4 (BPAI Aug. 10, 2009) (informative); see also 37
C.F.R. § 41.37(c)(iv) (“A statement which merely points out what a claim
recites will not be considered an argument for separate patentability of the
claim.”).
Claim 39 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 39 for the reasons
stated above.
Claim 28
Claim 28 recites that “the key-frame selector selects the key-frames
by determining an image quality for each candidate key-frame in each
cluster.” Appellant argues that image quality is defined as being “based on
the sharpness of the candidate key-frame or on the brightness of the
candidate key-frame or a combination of sharpness and brightness.” App.
Br. 13 (citing Spec. 23:17-20). However, the Specification recites that image
quality “may be” based on sharpness and brightness. Spec. 23:17. Thus, this
is permissive language regarding what image quality may be. Additionally,
Appeal 2010-005291
Application 10/489,076

10
the importation of limitations from an embodiment from the specification
into an independent claim which does not recite those limitations is
improper. See SuperGuide Corp. v. DirecTV Enters., Inc., 358 F.3d 870, 875
(Fed. Cir. 2004) (“For example, a particular embodiment appearing in the
written description may not be read into a claim when the claim language is
broader than the embodiment.”). Given that the Appellant chose broader
language representing an image quality rather than specifically reciting
“sharpness and brightness,” we find that the broadest reasonable
interpretation of “image quality” includes the motion of objects in the image
as cited by the Examiner. Ans. 13. Therefore, we are not persuaded by this
argument.
Claim 41 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 41 for the reasons
stated above.
Claim 30
Claim 30 recites that “the frame analyzers include a color layout
analyzer.” The Examiner relies on Toklu’s disclosure of a histogram
analyzer to show this claim limitation. Ans. 13. Appellant argues that color
layout is defined as “the color layout may be defined in terms of an average
color in each of a plurality of sub-blocks distributed across the image” as
opposed to a histogram which is “a representation of a frequency distribution
of colors in an image.” App. Br. 13 (citing Spec. 11:7-13-3). However, the
Specification recites that color layout “may be” defined in terms of an
average color. Spec. 11:11-32. Thus, this is permissive language regarding
what image quality may be. Additionally, the importation of limitations from
an embodiment from the specification into an independent claim which does
Appeal 2010-005291
Application 10/489,076

11
not recite those limitations is improper. See SuperGuide Corp. v. DirecTV
Enters., Inc., 358 F.3d 870, 875 (Fed. Cir. 2004) (“For example, a particular
embodiment appearing in the written description may not be read into a
claim when the claim language is broader than the embodiment.”). Given
that the Appellant chose the broader language “color layout” rather than
specifically reciting “an average color in each of a plurality of sub-blocks
distributed across the image,” we find that the broadest reasonable
interpretation of “color layout” includes a histogram. Ans. 14. Therefore, we
are not persuaded by this argument.
Claim 43 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 43 for the reasons
stated above.
Claim 31
Claim 31 recites that “the frame analyzers include a fast camera
motion detector.” Appellant argues this limitation is not met. . The
Specification states that “[t]he fast camera motion detector may detect a fast
motion of the camera that captured the video 12 by detecting a relatively
large difference in the color layouts or the color histograms of adjacent
video frames over a number of consecutive 20 video frames in the video
12.” Spec. 13:15-20. Thus, we find “fast motion” refers to detecting the
speed of the motion of the camera. Appellant admits that Toklu discloses
using thresholds to determine the speed of camera motions such as panning
and tilting. App. Br. 14 (citing Toklu, Fig 2B). Thus, we find that the motion
detection taught by Toklu, which can detect speed of camera motion, i.e. fast
motion, meets the fast camera motion detector in the context of the
invention. Therefore, we are not persuaded by this argument.
Appeal 2010-005291
Application 10/489,076

12
Claim 44 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 44 for the reasons
stated above.
Claim 33
Claim 33 recites that “the frame analyzers include an object motion
analyzer.” Appellant argues this limitation limited to specific camera
motions defined in the specification. App. Br. 15. However, these motions
are not recited in the claims, thus Appellant’s arguments are not
commensurate with the scope of the claim. Therefore, we are not persuaded
by this argument.
Claim 46 contains essentially the same limitations discussed above
and we affirm the Examiner’s decision to reject claim 46 for the reasons
stated above.

35 U.S.C. § 103(a) - Toklu and Wu
Claims 24, 27, 34, 37, 40, and 47
Level of Skill in the Art
Appellant argues that the Examiner must establish a level of ordinary
skill in the art regarding the Toklu and Wu references before finding that a
limitation would have been obvious to a person of ordinary skill in the art.
App. Br. 27-28. With regards to the level of ordinary skill in the art as noted
by our reviewing court, “the absence of specific findings on the level of skill
in the art does not give rise to reversible error ‘where the prior art itself
reflects an appropriate level and a need for testimony is not shown.”’
Appeal 2010-005291
Application 10/489,076

13
Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir. 2001). In cases where
the patents and prior art are “easily understandable, a factual determination
of the level of skill in the art was unnecessary. . . . [one] cannot, argue that
an unnecessary fact is material. It makes only a naked allegation that some
unspecified higher level of skill should have been applied.” Chrome-Time
Equipment Inc. v. Cumberland Corp., 713 F.2d 774, 779 (Fed. Cir.1983).
The Patent Office satisfies its initial burden of production by “adequately
explain[ing] the shortcomings it perceives so that the applicant is properly
notified and able to respond.” In re Jung, 637 F.3d 1356, 1362 (Fed. Cir.
2011) (citing Hyatt v. Dudas, 492 F.3d 1365, 1370 (Fed. Cir. 2007)). The
Examiner cited references in the art of video editing. We find that Toklu and
Wu reflect the appropriate level of skill in the art and Appellant has not
established otherwise with persuasive evidence.
Motivation to Combine
The Examiner explains that the motivation to combine Toklu with Wu
would be that “reduction in image processing would occur by including
audio analysis of Wu because frames with silence or noise would not be
processed. By eliminating those frames from processing, computational load
is reduced.” Ans. 15 (citing Wu at ¶ 0027). Appellant argues that “the image
processing reduction realized by Wu et al. is only associated with face
detection which Toklu et al. do [sic] not disclose using” App. Br. 29. We are
not persuaded by this argument.
We note that the U.S. Supreme Court has held that “[t]he obviousness
analysis cannot be confined by a formalistic conception of the words
teaching, suggestion, and motivation, or by overemphasis on the importance
Appeal 2010-005291
Application 10/489,076

14
of published articles and the explicit content of issued patents.” KSR Int’l
Co. v. Teleflex Inc., 550 U.S. 398, 419 (2007). Instead, the relevant inquiry
is whether the Examiner has set forth “some articulated reasoning with some
rational underpinning to support the legal conclusion of obviousness.” In re
Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006) (cited with approval in KSR, 550
U.S. at 418). “As our precedents make clear, however, the analysis need not
seek out precise teachings directed to the specific subject matter of the
challenged claim, for a court can take account of the inferences and creative
steps that a person of ordinary skill in the art would employ.” KSR, 550 U.S.
at 418.
Toklu teaches selecting key-frames based on a variety of analyses
which does not exclude facial recognition. Toklu, Fig. 1. Wu explicitly
teaches a method of avoiding unnecessary analysis of frames. Wu at ¶ 0027.
Thus, we agree with the Examiner that one of ordinary skill in the art would
be motivated to combine Wu’s frame analysis with Toklu’s frame analysis.
Ans. 15. Upon reviewing the record before us, we find that the Examiner’s
suggestion for modifying Toklu with Wu suffices as an articulated reason
with some rational underpinning to justify the legal conclusion of
obviousness.
Except as noted above, Appellant does not substantively argue the
rejection of claims 24, 27, 34, 37, 40, and 47. Thus, for the reasons stated
above, we affirm the Examiner’s decision to reject claims 24, 27, 34, 37, 40,
and 47.

Appeal 2010-005291
Application 10/489,076

15
35 U.S.C. § 103(a) - Toklu and Dimitrova
Claims 35 and 48
Level of Skill in the Art
Appellant argues that the Examiner must establish a level of ordinary
skill in the art regarding the Toklu and Dimitrova references before finding
that a limitation would have been obvious to a person of ordinary skill in the
art. App. Br. 32-33. We noted above that explicit establishment of a level of
ordinary skill is not required. We find that the cited Toklu and Dimitrova
references reflect the appropriate level of skill in the art and Appellant has
not established otherwise with persuasive evidence.
Motivation to Combine
Appellant argues that “Examiner's rationale for combining [Toklu and
Dimitrova] is essentially nothing more than a . . . conclusory statement”
App. Br. 30. The Examiner finds that “utilizing a user input would
predictably result an improved system by taking into account a user’s
desire.” Ans. 15.
While the Examiner must articulate some “reasoning with some
rational underpinning to support the legal conclusion of obviousness,” In re
Kahn, 441 F.3d 977, 988 (Fed. Cir 2006), in establishing that “there was an
apparent reason to combine the known elements in the fashion claimed,”
KSR Int’l Co. v. Teleflex, 550 U.S. 398, 418 (2007), the apparent reason
need not appear in, or be suggested by, one or more of the references on
which the Examiner relies. Instead, as noted above, the Examiner when
analyzing the evidence may employ common sense not inconsistent with the
ordinary level of knowledge and skill in the art. Perfect Web Techs. v.
Appeal 2010-005291
Application 10/489,076

16
InfoUSA, Inc., 587 F.3d 1324, 1328-29 (Fed. Cir. 2009). We find that the
combination of Toklu and Dimitrova involves the simple substitution of
Toklu’s automatic selection of frames with Dimitrova’s method of selecting
frames which is by user selection to yield predictable results. See KSR, 550
U.S. at 417. We agree with the Examiner that these teachings, all in the
context of selecting video frames, are properly combinable.
Except as noted above, Appellant does not substantively argue the
rejection of claims 35 and 48. Thus, for the reasons stated above, we affirm
the Examiner’s decision to reject claims 35 and 48.

DECISION
The Examiner’s decision to reject claims 1-5 and 10-12 is affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED

ELD