Adobe Inc.

Patent Trials and Appeals Board

Dec 6, 2021

2020006058 (P.T.A.B. Dec. 6, 2021)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
14/938,690 11/11/2015 Zeke Koch P2287-US 8959
108982 7590 12/06/2021
FIG. 1 Patents
116 W. Pacific Avenue
Suite 200
Spokane, WA 99201
EXAMINER
DAVANLOU, SOHEILA
ART UNIT PAPER NUMBER
2153
NOTIFICATION DATE DELIVERY MODE
12/06/2021 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
Fig1Docket@fig1patents.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte ZEKE KOCH, GAVIN STUART PETER MILLER,
JONATHAN W. BRANDT, NATHAN A. CARR, RADOMIR MECH,
WALTER WEI-TUH CHANG, SCOTT D. COHEN, and HAILIN JIN
___________

Appeal 2020-006058
Application 14/938,690
Technology Center 2100
__________

Before JAMES B. ARPIN, HUNG H. BUI, and SCOTT RAEVSKY,
Administrative Patent Judges.

ARPIN, Administrative Patent Judge.
DECISION ON APPEAL
Appellant1 appeals under 35 U.S.C. § 134(a) from the Examiner’s
decision rejecting claims 1–20, all of the pending claims. Final Act. 2.2 We
have jurisdiction under 35 U.S.C. § 6(b).
We affirm.

1 “Appellant” refers to “applicant” as defined in 37 C.F.R. § 1.42.
Appellant identifies the real party-in-interest as Adobe Inc. Appeal Br. 2.
2 In this Decision, we refer to Appellant’s Appeal Brief (“Appeal Br.,” filed
May 11, 2020) and Reply Brief (“Reply Br.,” filed August 12, 2020); the
Final Office Action (“Final Act.,” mailed December 23, 2019) and the
Examiner’s Answer (“Ans.,” mailed July 8, 2020); and the Specification
(“Spec.,” filed November 11, 2015). Rather than repeat the Examiner’s
findings and Appellant’s contentions in their entirety, we refer to these
documents.
Appeal 2020-006058
Application 14/938,690

2

STATEMENT OF THE CASE
The claimed methods and systems relate to “[c]ontent creation
collection and navigation techniques and systems.” Spec. ¶ 4.
As noted above, claims 1–20 are pending. Claims 1, 9, and 15 are
independent. Appeal Br. 27 (claim 1), 29 (claim 9), 31 (claim 15) (Claims
App.). Claims 2–8 depend directly or indirectly from claim 1, claims 10–14
depend directly or indirectly from claim 9, and claims 16–20 depend directly
or indirectly from claim 15. Id. at 27–32.
Claims 1 and 15, reproduced below with the disputed limitations
emphasized, are illustrative.
1. In a digital medium environment for image search
result configuration and navigation control, a method
implemented by one or more computing devices comprising:
obtaining, by the one or more computing devices, a search
result including a plurality of licensable images available for
licensing from a content sharing service, the plurality of
licensable images identified by the content sharing service in
response to a search request;
generating, by the one or more computing devices,
clusters of the plurality of licensable images based on a first
metric to measure similarity of the plurality of licensable images,
the clusters including images determined to be similar based on
the first metric;
identifying, by the one or more computing devices and
after generating the clusters, at least a second metric to
differentiate between clustered images of an individual cluster
that have previously been clustered based on the first metric;
determining, by the one or more computing devices, a
representative licensable image for each of the clusters;
generating, by the one or more computing devices, a user
interface to display the representative licensable images of the
clusters and an image navigation control, the image navigation
Appeal 2020-006058
Application 14/938,690

3

control supporting user navigation through the licensable
images of a selected cluster based on the at least second metric
and responsive to a user selection of the representative
licensable image of the selected cluster;
receiving, by the one or more computing devices, a request
initiated via the user interface to license at least one of the
licensable images; and
licensing, by the one or more computing devices, the
requested licensable image for use.
Id. at 27 (emphases added). Independent claim 9 recites limitations
corresponding to the disputed limitations of claim 1. Id. at 29; see Ans. 6;
Reply Br. 6.
15. In a digital medium environment for image sharing, a
method implemented by one or more computing devices of a
content sharing service comprising:
receiving a plurality of images by the one or more
computing devices of the content sharing service;
grouping the plurality of images responsive to a
determination by the one or more computing devices that the
plurality of images exhibit at least a threshold amount of
similarity, one to another;
identifying, after the grouping of the plurality of images,
one or more differentiation metrics that differentiate between the
similarity exhibited by the plurality of images previously
grouped based on the threshold amount of similarity; and
responsive to locating at least one of the plurality of
images as corresponding to a search request received by the
content sharing service, configuring a search result by the one or
more computing devices that includes the at least one said image
as representative of a grouping as a whole in the search result,
and further configuring the search result with functionality that
is selectable to cause output of one or more other said images of
the grouping and to enable user navigation through the grouping,
the one or more other said images not being output absent the
selection, and the user navigation through the grouping enabled
Appeal 2020-006058
Application 14/938,690

4

based on the one or more differentiation metrics to differentiate
between the similarity exhibited by the one or more images of
the grouping.
Id. at 31 (emphases added).
REFERENCES AND REJECTIONS
The Examiner relies upon the following references:
Name3 Reference Publ’d/Issued Filed
Ciorba US 9,219,830 B1 Dec. 22, 2015 Apr. 3, 2015
Moseley US 2013/0246277 A1 Sept. 19, 2013 Mar. 13, 2012
Howard US 2014/0108931 A1 Apr. 17, 2014 Dec. 30, 2013
The Examiner rejects: (1) claims 1–14 under 35 U.S.C. § 103 as
obvious over the combined teachings of Howard, Ciorba, and Moseley
(Final Act. 2–16); and (2) claims 15–20 under 35 U.S.C. § 103 as obvious
over the combined teachings of Howard and Ciorba (id. at 16–24).
We review the appealed rejections for error based upon the issues
identified by Appellant, and in light of the contentions and evidence
produced thereon. Ex parte Frye, 94 USPQ2d 1072, 1075 (BPAI 2010)
(precedential). The Examiner and Appellant focus their findings and
contentions, respectively, on the disputed limitations of independent
claims 1, 9, and 15 (see Appeal Br. 8, 16, 22; Ans. 3, 6, 7); so we do as well.
Arguments not made are forfeited.4 Unless otherwise indicated, we adopt

3 All reference citations are to the first named inventor only.
4 See In re Google Tech. Holdings LLC, 980 F.3d 858, 863 (Fed. Cir. 2020)
(“Because Google failed to present these claim construction arguments to
the Board, Google forfeited both arguments.”); 37 C.F.R. § 41.37(c)(1)(iv)
(2019) (“Except as provided for in §§ 41.41, 41.47 and 41.52, any arguments
Appeal 2020-006058
Application 14/938,690

5

the Examiner’s findings in the Final Office Action and the Answer as our
own and add any additional findings of fact for emphasis. We address the
rejections below.
ANALYSIS
A. Obviousness over Howard, Ciorba, and Moseley
As noted above, the Examiner rejects claims 1–14 under 35 U.S.C.
§ 103 as obvious over the combined teachings of Howard, Ciorba, and
Moseley. Final Act. 2–16. In particular, the Examiner finds Howard
teaches or suggests the majority of the limitations recited in independent
claim 1. Id. at 2–4 (citing Howard ¶¶ 9, 19, 32, 34, 35, Figs. 2B–2D).
Nevertheless, the Examiner finds, Howard does not specifically teach,
“identifying, by the one or more computing devices and after generating the
clusters, at least a second metric to differentiate between clustered images of
an individual cluster that have previously been clustered based on the first
metric.” Id. at 4; see Appeal Br. 27 (Claims App.). The Examiner finds,
however, that Ciorba teaches or suggests this limitation (Final Act. 4–5
(citing Ciorba, 27:26–35); see Ciorba, 9:18–20 (defining “Saliency”)) and
that a person of ordinary skill in the relevant art would have had reason to
combine the teachings of Howard and Ciorba to achieve this limitation
(Final Act. 5 (citing Ciorba, 2:32–35)).

or authorities not included in the appeal brief will be refused consideration
by the Board for purposes of the present appeal.”).
Appeal 2020-006058
Application 14/938,690

6

In addition, the Examiner finds Howard and Ciorba together teach or
suggest,
determining, by the one or more computing devices, a
representative licensable image for each of the clusters;
generating, by the one or more computing devices, a user
interface to display the representative licensable images of the
clusters and an image navigation control, the image navigation
control supporting user navigation through the licensable images
of a selected cluster based on the at least second metric and
responsive to a user selection of the representative licensable
image of the selected cluster.
Final Act. 5–7 (citing Howard ¶¶ 32, 35, 36, Figs. 2A–2M and Ciorba,
11:15–23, 30:14–16). The reason to combine Howard’s teachings with
those of Ciorba, cited above, applies equally here. See Ciorba, 2:32–35.
The Examiner further finds that Howard and Ciorba fail to teach or
suggest,
receiving, by the one or more computing devices, a request
initiated via the user interface to license at least one of the
licensable images; and
licensing, by the one or more computing devices, the
requested licensable image for use.
Final Act. 7; see Appeal Br. 27 (Claims App.). Nevertheless, the Examiner
finds that Moseley teaches or suggests these missing limitations (Final Act.
7–8 (citing Moseley ¶ 26, Fig. 2)) and that a person of ordinary skill in the
relevant art would have had reason to combine the teachings of Howard and
Ciorba with those of Moseley to achieve the methods of claim 1 (id. at 8
(citing Moseley ¶ 7)).
Appellant contends the Examiner errs in rejecting independent
claim 1 for three reasons. Appeal Br. 8–15. On this record, Appellant does
not persuade us the Examiner errs.
Appeal 2020-006058
Application 14/938,690

7

1. “generating” limitation
First, independent claim 1 recites, “generating . . . clusters of the
plurality of licensable images based on a first metric to measure similarity
of the plurality of licensable images, the clusters including images
determined to be similar based on the first metric.” Appeal Br. 27 (Claims
App.) (emphasis added). Appellant contends, “the Examiner asserts Howard
for generating ‘clusters of the plurality of licensable images.’ However,
Howard is missing generating ‘clusters of the plurality of licensable images,’
in the particular manner claimed.” Id. at 8. For the reasons given below, we
disagree with Appellant.
The Examiner finds Howard discloses, “[s]earch base modules may
include account-based modules such as Flickr 202, mentioned above, and
Picasa 203 that present an authentication interface allowing a user to connect
to the user’s public and/or private photos at a third party hosting service.”
Howard ¶ 34; see Final Act. 3; Ans. 4; see also Howard, Fig. 2E (including
“media credits” with license information). Thus, the Examiner finds
Howard teaches or suggests clusters of the plurality of licensable images
based on a first metric, for example, a search term, such as “sunflowers.”
See Final Act. 3 (citing Howard ¶ 35, Figs. 2B–2D).
Appellant contends, however, that
Howard merely describes clustering images that match a
particular search query. The asserted portion of Howard, for
example, describes that “typing in ‘sunflowers’ in a Flickr search
query results in pictures of sunflowers or photos from Flickr
being presented on the user’s [graphical user interface (GUI)]”
such that the pictures of sunflowers may be viewed “as photo
clusters.”
Appeal Br. 9 (quoting Howard ¶ 35). Appellant further contends:
Appeal 2020-006058
Application 14/938,690

8

Matching a search term is not a similarity measure
though. Indeed, matching a textual search term by two images is
not the same as a metric to measure similarity between two
different images, such that the two different images may be
determined to be similar or dissimilar based on the metric.
Id. (emphasis added); see Reply Br. 3. We disagree with Appellant.
The Specification discloses:
In the illustrated example, images 206, 208, 210 exhibit
similarity as including matching objects, with differences being
movement of an object (e.g., a basketball) from one image to
another. This similarity may be determined in a variety of ways,
such as through use of predefined metrics including color,
contrast, saturation, use of similar image processing as part of a
content creation service 102. . . . In another example, the
similarity determination is made through comparison of the
images, one to another, such as to determine whether the
images 204 have similar objects, perspectives, lighting, and so
forth and thus exhibit at least at threshold amount of similarity,
one to another. Other examples are also contemplated. . . . Real
time examples are also contemplated in which this determination
is performed as part of processing of a search result generated
in response to a search request.
Spec. ¶ 35 (emphases added). Thus, we understand the recited “first metric
to measure similarity” broadly encompasses a variety of metrics including a
search request for similar objects based on a search term, such as
“sunflower.” See id., Figs. 2 (images 206, 208, and 210 of a “basketball”),
4 (images 420, 422, 424, and 426 of a “man” with a “briefcase”); see also id.
¶ 39 (“For instance, a user may perform a search for a ‘businessman’ for use
in marketing materials and view image 414.”).
Appellant also contends, “[t]here is no mention in Howard of similar
images or metrics to measure similarity between images.” Appeal Br. 9.
However, Howard’s Figures 2B and 2C depict similar images of
Appeal 2020-006058
Application 14/938,690

9

“sunflowers” collected from different sources, i.e., Flickr and Google.
Howard ¶ 35 (“A result of the aforementioned search query in Flickr is
shown in FIG. 2B. The collected media (sunflowers) may be viewed
individually as shown in FIG. 2C or as photo clusters as shown in
FIG. 2D.”). Moreover, those images may be clustered as depicted in
Howard’s Figure 2D. Id. Thus, Appellant does not persuade us that Howard
fails to teach or suggest this disputed limitation.
Consequently, we are not persuaded of Examiner error by this first
reason.
2. “identifying” limitation
Second, independent claim 1 recites, “identifying, by the one or more
computing devices and after generating the clusters, at least a second metric
to differentiate between clustered images of an individual cluster that have
previously been clustered based on the first metric.” Appeal Br. 27 (Claims
App.) (emphasis added). In particular, Appellant contends Ciorba fails to
teach or suggest, “identifying . . . at least a second metric.” Id. at 11. For
the reasons given below, we disagree with Appellant.
Initially, Appellant contends, “Ciorba does not first cluster images
according to a measure of similarity and then identify a second metric to
differentiate between the images in a given cluster of ‘images determined to
be similar based on the first metric . . . .’” Id. (emphasis added). As noted
above, however, the Examiner relies on Howard, not Ciorba, to teach or
suggest the emphasized limitation. See supra Section A.1. Appellant cannot
show nonobviousness by attacking references individually when the
rejection is based on the references’ combined teachings. See In re Merck &
Co., Inc., 800 F.2d 1091, 1097 (Fed. Cir. 1986); In re Keller, 642 F.2d 413,
Appeal 2020-006058
Application 14/938,690

10

426 (CCPA 1981). Thus, this contention is not persuasive of Examiner
error.
The Examiner further finds that:
As shown above Howard teaches showing clusters of
images of the individual cluster (e.g. sunflowers) to differentiate
between clustered images based on first metric, Howard does not
specifically teach identifying after the generating of the clusters.
In addition Ciorba in col. 27, lines 26-35 teaches at least a
second metric [Saliency] to differentiate between clustered
images of the plurality of licensable images, the at least second
metric differentiating between the clustered images of an
individual cluster (e.g. specific heuristics image analysis and
saliency analysis) that have previously been clustered based on
the first metric (e.g. similarity of images).
Ans. 4–5. In particular, the Examiner finds the first metric, used to gather
and cluster images, may be a search term or an author and time similarity
(Howard ¶¶ 35, 37) and the second metric may be “use-case specific
heuristics” (Ciorba, 27:26–35). See Ciorba, 9:18–20 (“‘Saliency’ refers to a
most relevant and/or most important part of an image as determined by
image analysis or user action analysis (user inputs).”). Specifically, Ciorba
discloses such heuristics may comprise “grouping images having identical
or similar faces (based on the faces that were previously identified),
grouping images having identical objects (based on the objects that were
previously identified).” Id. at 27:26–31 (emphases added). Thus, the
Examiner finds that Howard and Ciorba together teach or suggest gathering
and clustering images based on a first metric and differentiating previously
clustered images based on a second metric. See Final Act. 4–5, 24–25. We
agree with the Examiner. Therefore, Appellant does not persuade us that
Ciorba in view of Howard fails to teach or suggest this disputed limitation.
Appeal 2020-006058
Application 14/938,690

11

Consequently, we are not persuaded of Examiner error by this second
reason.
3. “image navigation control” limitation
Third, independent claim 1 recites, “an image navigation control, the
image navigation control supporting user navigation through the licensable
images of a selected cluster based on the at least second metric and
responsive to a user selection of the representative licensable image of the
selected cluster.” Appeal Br. 27 (Claims App.) (emphases added). In
particular, Appellant contends,
the Examiner argues that Ciorba’s “dragging and dropping of
photos” is user navigation through a plurality of licensable
images. This is incorrect, however. Dragging or dropping is
operating on or moving photos. It is not navigating through
images, e.g., to view them. Additionally, the Examiner
completely divorces the claimed “image navigation control”
from the feature that the navigation supported by the control is
“based on the at least second metric.” Indeed, dragging and
dropping is user based. In other words, dragging and dropping
is based on a user’s decision to select a photo, then drag the photo
to a location determined by the user, and then drop it. No
identification of a second metric is involved in enabling such user
interaction.
Id. at 12–13 (emphases added). Moreover, Appellant contends that, unlike
Ciorba, slider 708, depicted in Appellant’s Figure 7, shows navigation
through images ordered based on a “meaningful metric,” as recited in
claim 1. Id. at 14–15. We disagree with Appellant.
The Examiner finds that “Howard in paragraph [0035] teaches search
query results in pictures of sunflowers or photos presented on the user’s GUI
or media page and may be viewed (e.g. navigate) individually as shown in
FIG. 2C or as photo clusters as shown in FIG. 2D.” Ans. 5; see Final
Appeal 2020-006058
Application 14/938,690

12

Act. 3–4. Further, the Examiner finds Ciorba teaches identifying a second
metric to differentiate between the clustered images (see supra Section A.2.)
and
the image navigation control supporting user navigation [e.g.
dragging and dropping of photos] through the licensable
images [e.g. applied image analysis on analyzed photos] (Ciorba,
col. 11, lines 15-23, photo editing or correction may be
automatically performed by using an image analysis engine that
is capable of extracting photo metadata and applying image
heuristics/use cases on analyzed photos . . . a photobook is
produced by dragging and dropping of photos) of a selected
cluster based on the at least second metric [e.g. analyzed
detecting areas of saliency, face detection] (Ciorba, col. 30, lines
14-16, after using the automatic image information derived by
image analysis (detecting areas of saliency, face detection), the
server outputs a default design preview)
(Final Act. 6–7 (italics added)). Thus, the Examiner finds after an initial
clustering based on a first metric, e.g., a search term, as taught by Howard,
the clustered images may be further clustered and navigated, e.g., analyzed
for saliency (e.g., a second metric) and then may be dragged and dropped
based on that analysis. Moreover, we understand the “selected cluster [is]
based on the at least second metric,” not the image navigation control.
Appeal Br. 27 (Claims App.) (emphases added); Spec. ¶ 39 (“Through
selection of the representation of this image 414, however, images having
different orientations of the businessman are output and selectable by a user
for licensing and thus a user may then navigate [these] similar images to
locate one of interest.”). Therefore, the Examiner finds Howard and Ciorba
together teach or suggest this disputed limitation. Final Act. 6–7.
In response to Appellant’s contentions regarding the disclosure of
Appellant’s Figure 7, the Examiner further finds, “[w]hile appreciated that
Appeal 2020-006058
Application 14/938,690

13

the appellant specification’s has a particular slider to navigate through the
various clustered images (see Fig. 7), the claimed user navigation merely
recites image navigation control, in other words, which is given [its] plain
meaning definition of a interface control to navigate images.” Ans. 5. We
agree with the Examiner.
Although we may determine the scope of the claims in patent
applications not solely on the basis of the claim language, we are careful not
to read a particular embodiment appearing in the specification into the claim
when the claim language is broader than the embodiment. See SuperGuide
Corp. v. DirecTV Enterprises, 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.”). Here, the claim language does not describe the structure of
an image navigation control, only that the control functions to “support[]
user navigation through the licensable images of a selected cluster based on
the at least second metric,” and the Specification does not limit the control
to that depicted in Appellant’s Figure 7. See Appeal Br. 27 (Claims App.)
(emphases added); Spec. ¶¶ 9 (“FIG. 7 depicts an example implementation
of user interfaces in which a user interface image navigation control is
configured to support user navigation through images based on one or more
metrics.” (emphasis added)); 49 (“Additional examples [to that depicted in
Figure 7] are also contemplated as described in further detail below.”).
Therefore, we are not persuaded the broad language of claim 1 should be
limited to the particular embodiment of Figure 7, and the absence of
slider 708 or a similar control from Ciorba does not distinguish this disputed
limitation over the combined teachings of Howard and Ciorba.
Appeal 2020-006058
Application 14/938,690

14

Consequently, we are not persuaded of Examiner error by this third
reason.
On this record, we are not persuaded the Examiner errs in rejecting
independent claim 1 as obvious over the combined teachings of Howard,
Ciorba, and Moseley. As noted above, claim 9 recites limitations
corresponding to the disputed limitations of claim 1, and Appellant
challenges the rejection of independent claim 9 for substantially the same
reasons as claim 1. See Ans. 6; Reply Br. 6. Thus, we sustain the rejection
of claims 1 and 9. Appellant does not challenge the obviousness rejection of
dependent claims 2–8 and 10–14 separately from their base claims. See
Appeal Br. 14. Therefore, we also sustain the rejection of those claims.
B. Obviousness over Howard and Ciorba
As noted above, the Examiner rejects claims 15–20 under 35 U.S.C.
§ 103 as obvious over the combined teachings of Howard and Ciorba. Final
Act. 16–24. In particular, the Examiner finds Howard teaches or suggests
the majority of the limitations recited in independent claim 15. Id. at 16–18
(citing Howard ¶¶ 19, 34–37, Figs. 2B–2E). Nevertheless, the Examiner
finds, Howard does not specifically teach,
identifying after the grouping of the plurality of images
one or more differentiation metrics that differentiate between the
similarity exhibited by the plurality of images previously
grouped based on the threshold amount of similarity; and
the one or more other said images not being output absent
the selection the user navigation through the grouping enabled
based on the one or more differentiation metrics to differentiate
between the similarity exhibited by the one or more images of
the grouping.
Appeal 2020-006058
Application 14/938,690

15

Id. at 18–19; see Appeal Br. 31 (Claims App.). The Examiner finds,
however, that Ciorba teaches or suggests these missing limitations (Final
Act. 19–20 (citing Ciorba, 9:18–20, 11:15–23, 27:26–35)) and that a person
of ordinary skill in the relevant art would have had reason to combine the
teachings of Howard and Ciorba to achieve the methods of claim 15 (id. at
20–21 (citing Ciorba, 2:32–35)).
Appellant contends the Examiner errs in rejecting independent
claim 15 for two reasons. Appeal Br. 22–25. On this record, Appellant does
not persuade us the Examiner errs.
1. “grouping” limitation
First, independent claim 15 recites, “grouping the plurality of images
responsive to a determination by the one or more computing devices that the
plurality of images exhibit at least a threshold amount of similarity, one to
another.” Appeal Br. 31 (Claims App.) (emphases added). Appellant
contends:
Matching a search term, as described in Howard, is not a
threshold amount of similarity though. In addition, the mere
presentation of pictures that match a search term (e.g.,
“sunflowers”) in clusters, as described in Howard, does not
correspond to “grouping the plurality of images responsive to a
determination by the one or more computing devices that the
plurality of images exhibit at least a threshold amount of
similarity, one to another,” as recited in claim 15. This is because
matching or not matching a search term is different from images
having at least a threshold amount of similarity between them.
Moreover, the only mention of the term “similar” or “similarity”
in Howard pertains to “similar search services.” There is no
mention in Howard of similar images or amounts of similarity
between images.
Appeal 2020-006058
Application 14/938,690

16

Id. at 23 (emphasis added; footnote omitted; citing Howard ¶ 44). For the
reasons given below, we disagree with Appellant.
The Examiner finds, “Howard in paragraph [0037] teaches the further
images clustering similarity (time and author) collected within a
predetermined time frame (e.g. threshold amount of similarity) by the same
author may be clustered together.” Ans. 7; see Final Act. 16–17 (“As such,
in one embodiment, those photos collected within a predetermined time
frame by the same author may be clustered together (threshold amount of
similarity).”). In particular, Howard discloses:
If consecutive photos fall within the window they are clustered
together. The window duration is chosen based on a heuristic,
generally, between for example, 10 seconds and 15 minutes are
use based on the search provider. For example, as shown in
FIG. 2E, each media item includes credits such as date created,
source (Flickr.com), and author (Sankova.O). As such, in one
embodiment, those photos collected within a predetermined time
frame by the same author may be clustered together.
Howard ¶ 37 (emphases added). Thus, the Examiner finds that images
obtained by the same search term based on a request from the same source
and created by the same author and collected within a particular time frame
teach or suggest, “exhibit[ing] at least a threshold amount of similarity, one
to another.” Ans. 7; Appeal Br. 31 (Claims App.).
Appellant responds that
a window of time (e.g., predetermined or otherwise) as described
in Howard is not a “threshold amount of similarity” as claimed.
This is at least in part because a window of time does not provide
a basis for an “amount of similarity,” such that two images may
be more or less similar depending on the amount of similarity.
Reply Br. 7 (emphasis added). A reference, however, “must be considered
for all that it teaches to those of ordinary skill in the art, not just the
Appeal 2020-006058
Application 14/938,690

17

embodiments disclosed therein.” In re Arora, 369 F. App’x 120, 122 (Fed.
Cir. 2010); see In re Inland Steel Co., 265 F.3d 1354, 1361 (Fed. Cir. 2001);
In re Fritch, 972 F.2d 1260, 1264 (Fed. Cir. 1992). Here, the Examiner
relies on all that Howard teaches in paragraph 37, specifically, the use of the
same search term, the same source, the same author, and the limited time
frame to teach or suggest the threshold amount of similarity. Ans. 7. We
are not persuaded the Examiner errs in finding Howard teaches or suggests
this disputed limitation.
Consequently, we are not persuaded of Examiner error by this first
reason.
2. “identifying” limitation
Second, independent claim 15 recites, “identifying, after the grouping
of the plurality of images, one or more differentiation metrics that
differentiate between the similarity exhibited by the plurality of images
previously grouped based on the threshold amount of similarity.” Appeal
Br. 31 (Claims App.). Appellant contends,
the Examiner asserts that “grouping images having identical
objects (based on the objects that were previously identified)”
corresponds to a second metric. Initially, it is asserted that
identification of an object is not identification of “a second
metric.” The mere use of different known aspects of images to
group images does not correspond to “identifying” metrics.
Id. at 24 (footnote omitted; quoting Final Act. 19). For the reasons given
below, we disagree with Appellant.
The Specification discloses:
A meaningful metric is also identified to differentiate the
images 710, 712, 714, which in this case is orientation of the
object in the images. Accordingly, the images 710, 712, 714 are
Appeal 2020-006058
Application 14/938,690

18

then ordered based on this identified meaningful metric (e.g.,
similarity of orientations, one to another) and the slider 708 is
used to navigate through this order. In this example, the
meaningful metric (e.g., orientation) is chosen to differentiate
the basis of the similarity determination (e.g., inclusion of a
particular object), although other examples are also
contemplated. In this way, similar images may be efficiently
located and the meaningful metric utilized to navigate through
differences between these similar images. Additional examples
are also contemplated as described in further detail below.
Spec. ¶ 49 (emphases added). Thus, although “orientation” may be used to
differentiate between similar images, other metrics may be used to
differentiate between similar images. See id. ¶ 53 (“In this way, rather than
solely rely on a user’s ability to arrive at keywords that accurately reflect the
user’s desires, additional information from the images themselves is
leveraged as part of the search.”).
The Examiner finds that Howard teaches grouping clusters of images
based on a search term, i.e., sunflowers. Ans. 7. This clustering is based on
previously identified objects, such as the search term. Id.; see Final Act. 16–
17 (“For example, selecting the ‘full text’ search criteria and typing in
‘sunflowers’ in a Flickr search query results in pictures of sunflowers or
photos from Flickr being presented on the user’s GUI or media page.”).
Further, the Examiner finds,
Ciorba in col. 27, lines 26-35 teaches at least a second metric
(e.g. Saliency) to differentiate between clustered images of the
plurality of licensable images, the at least second metric
differentiating between the clustered images of an individual
cluster (specific heuristics image analysis and saliency analysis)
that have previously been clustered based on the first metric (e.g.
similarity of images).
Ans. 7–8; see Final Act. 19. Specifically, Ciorba discloses
Appeal 2020-006058
Application 14/938,690

19

The use-case specific heuristics comprise: grouping images that
have previously been found to be similar, grouping images
having identical or similar faces (based on the faces that were
previously identified), grouping images having identical objects
(based on the objects that were previously identified) . . . .
Ciorba, 27:26–31 (emphases added); see id. at 9:18–20 (defining
“Saliency”). Thus, the Examiner finds that Howard and Ciorba together
teach or suggest initially grouping images based on a threshold amount of
similarity and then identifying one or more differentiation metrics, such as
identical or similar faces or identical objects, to differentiate between the
previously grouped images. See Final Act. 20. We agree with the
Examiner.5
Consequently, we are not persuaded of Examiner error by this second
reason.
On this record, we are not persuaded the Examiner errs in rejecting
independent claim 15 as obvious over the combined teachings of Howard
and Ciorba. Thus, we sustain the rejection of claim 15. Appellant does not
challenge the obviousness rejection of dependent claims 16–20 separately.
See Appeal Br. 15, 22. Therefore, we also sustain the rejection of those
claims.
DECISION
1. The Examiner does not err in rejecting:
a. claims 1–14 under 35 U.S.C. § 103 as obvious over the
combined teachings of Howard, Ciorba, and Moseley; and

5 In the Reply Brief, Appellant does not address the Examiner’s response
(Ans. 7–8) to Appellant’s contentions regarding this “identifying” limitation
(Appeal Br. 24–25).
Appeal 2020-006058
Application 14/938,690

20

b. claims 15–20 under 35 U.S.C. § 103 as obvious over the
combined teachings of Howard and Ciorba.
2. Thus, on this record, claims 1–20 are not patentable.
CONCLUSION
We affirm the Examiner’s rejections of claims 1–20.
DECISION SUMMARY
In summary:
Claim(s)
Rejected
35
U.S.C. §
Reference(s)/Basis Affirmed Reversed
1–14 103 Howard, Ciorba,
Moseley
1–14
15–20 103 Howard, Ciorba 15–20
Overall
Outcome
1–20
TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).
AFFIRMED