Ex Parte Liu et al

Patent Trial and Appeal Board

Jan 22, 2018

14092210 (P.T.A.B. Jan. 22, 2018)

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.
14/092,210 11/27/2013 Siying Liu 3718US01 3167
108982 7590
Wolfe-SBMC
116 W. Pacific Avenue
Suite 200
Spokane, WA 99201
EXAMINER
NGUYEN, VU
ART UNIT PAPER NUMBER
2619
NOTIFICATION DATE DELIVERY MODE
01/24/2018 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):
docket@sbmc-law.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte SIYING LIU, KALYAN SUNKAVALLI, NATHAN A. CARR,
and ELYA SHECHTMAN
Appeal 2017-008227
Application 14/092,2101
Technology Center 2600
Before DENISE M. POTHIER, SCOTT B. HOWARD, and
JOYCE CRAIG, Administrative Patent Judges.
HOWARD, Administrative Patent Judge.
DECISION ON APPEAL
Appellant appeals under 35 U.S.C. § 134(a) from the Final Rejection
of claims 1-9 and 11-20, which constitute all of the claims rejected in this
application. Claim 10 is cancelled, and claim 21 is objected to as being
dependent on a rejected based claim. Br. 36; Final Act. 2. We have
jurisdiction under 35 U.S.C. § 6(b).
We affirm.
1 Appellant is the Applicant, Adobe Systems Incorporated, which is
identified in the Appeal Brief as the real party in interest. Br. 3.
Appeal 2017-008227
Application 14/092,210
THE INVENTION
The disclosed and claimed invention is directed to a texture modeling
of image data. Abstract. More specifically, according to Appellant, “texels
in image data are discovered by one or more computing devices, each texel
representing an element that repeats to form a texture pattern in the image
data.” Id. Additionally, the “[regularity of the texels in the image data is
modeled by the one or more computing devices to define translations and at
least one other transformation of texels in relation to each other.” Id.
Claim 1, reproduced below, is illustrative of the claimed subject
matter:
1. A method comprising:
discovering texels in image data by one or more
computing devices, each of the texels representing an element
that repeats to form a texture pattern in the image
data;
modeling regularity of the texels in the image data by the
one or more computing devices to form a texture model
describing a spatial configuration and arrangement of the texels
at a texel level, the texture model defining translations and at
least one other transformation of texels in relation to each other;
and
performing one or more texture manipulation editing
operations at the texel level based on the texture model.
REFERENCES
The prior art relied upon by the Examiner as evidence in rejecting the
claims on appeal is:
Racape US 2014/006463 0 A1 Mar. 6,2014
Connelly Barnes et al., PatchMatch: A Randomized Correspondence
Algorithm for Structural Image Editing. 28 ACM Transactions on
Graphics (Proc. SIGGRAPH) 24:1-11 (2009) (hereinafter “Shechtman”).
2
Appeal 2017-008227
Application 14/092,210
Connelly Bames et al., The Generalized PatchMatch Correspondence
Algorithm, European Conference on Computer Vision 1-14 (Sept. 2010)
(hereinafter “Bames”).
REJECTIONS
Claims 1-9 and 11-20 stand rejected under 35 U.S.C. § 103 as being
unpatentable over Bames in view of Shechtman and Racape. Final Act. 3-
10.
ANALYSIS
We have reviewed the Examiner’s rejection in light of Appellant’s
arguments that the Examiner erred. In reaching this decision, we have
considered all evidence presented and all arguments made by Appellant. We
are not persuaded by Appellant’s arguments regarding claims 1-9 and 11-
20, and we incorporate herein and adopt as our own: (1) the findings and
reasons set forth by the Examiner in the action from which this appeal is
taken (Final Act. 3-10), and (2) the reasons and rebuttals set forth in the
Examiner’s Answer in response to Appellant’s arguments (Ans. 10-14). We
incorporate such findings, reasons, and rebuttals herein by reference unless
otherwise noted. However, we highlight and address specific findings and
arguments for emphasis as follows.
Claims 19
Appellant argues the Examiner erred in finding the combination of
Bames and Shechtman teaches the “modeling regularity of the texels in the
image date” step recited in claim 1. Br. 11-21. Specifically, Appellant
argues “Bames describes an algorithm for discovering repeating texels
3
Appeal 2017-008227
Application 14/092,210
and/or repeated elements in an image, Barnes stops at this discovery.” Br.
13. Therefore, according to Appellant, “Barnes does not [teach] ‘modeling
regularity of the texels in the image data,’ ‘forming a texture model,’ nor
[sic] ‘defining translations and at least one other transformation of texels in
relation to each other’ as recited by claim 1.” Id. As Appellant explains:
For example, in an image containing repeating textures,
some of which are rotated 90 degrees, the generalized
PatchMatch algorithm can discover these repeating textures
regardless of the 90-degree rotation (e.g., the generalized
PatchMatch algorithm can discover repeating textures A1 and
A2 in the image data, even if A2 is rotated 90 degrees in
relation to Al). However, as described by Barnes, the
generalized PatchMatch algorithm does not and would not
provide any indication that A2 is rotated nor does the
generalized PatchMatch algorithm define in any way how A2 is
rotated in relation to Al (e.g., A2 is rotated 90 degrees to the
right in relation to Al) nor translation of these textures. At
most, as described by Barnes, the generalized PatchMatch
algorithm would merely indicate that Al and A2 are matching,
repeating textures, without defining or indicating their
arrangement, relational translation, or relational transformation
in any way.
Id. Appellant makes a similar argument regarding the related Shechtman
publication. Id. at 19-20. In other words, Appellant argues that although
Barnes and Shechtman teach discovering texels (repeating textures in an
image), the claim requires more.
The Examiner concludes the broadest reasonable construction of
“modeling” is broad enough to encompass the patch matching routines of
Barnes:
For broadest reasonable interpretation purposes and
without importing limitations from the specification into the
claim (MPEP §§2111 and 2111.01), the [Specification
discloses that “[t]o discover self-similar patterns within the
4
Appeal 2017-008227
Application 14/092,210
image, a patch matching technique may be enhanced to find k-
nearest neighbor (k-NN) patches across scales (e.g., resizing),
rotations, skews, and perspective distortions. An example of a
patch matching technique is described as The Generalized
PatchMatch Correspondence Algorithm, by C. Barnes, E.
Shechtman, A. Finkelstein, and D. B. Goldman, In European
Conference on Computer Vision, September 2010, the
disclosure of which is_hereby incorporated by reference in its
entirety.” PGPUB, [0021], This is the Barnes reference in
the cited Bames-Shechtman-Racape combination.
Additionally, the ‘other transformations’ recited in the
limitation can be, among other things, “scales, rotations, skews,
or perspective distortions.” PGPUB, [0065].
Ans. 11.
Applying that claim construction, the Examiner finds the combination
of Barnes, Shechtman, and Racape teaches “that the patch matching
technique (same algorithm used by the claimed invention) operate on image
descriptors such as SIFT,” which “matches patches across translations,
rotations, and scales.” Id. at 12 (citing Barnes 2:60-63, 3:62-67). The
Examiner further finds “the same patch matching algorithm used in the
Bames-Shechtman-Racape combination and the claimed invention
inherently produces a result (i.e. model) that define how the discovered
texture pattem/self-similar patterns are disposed within an image in relation
to each other across translations and at least one other transformation.” Id.;
see also Bames 11:53-63, Fig. 6.
We are not persuaded of error based on Appellant’s argument because
it does not address the conclusions, findings, and reasoning relied on by the
Examiner in the Answer and, thus, does not adequately address the rejection
on appeal. Instead, we agree with the Examiner in determining the above
combination teaches this disputed element.
5
Appeal 2017-008227
Application 14/092,210
Specifically, during examination of a patent application, a claim is
given its broadest reasonable construction “in light of the specification as it
would be interpreted by one of ordinary skill in the art.” In re Am. Acad, of
Sci. Tech Ctr., 367 F.3d 1359, 1364 (Fed. Cir. 2004) (quoting In re Bond,
910 F.2d 831, 833 (Fed. Cir. 1990). Under this standard, we interpret claim
terms using “the broadest reasonable meaning of the words in their ordinary
usage as they would be understood by one of ordinary skill in the art, taking
into account whatever enlightenment by way of definitions or otherwise that
may be afforded by the written description contained in the applicant’s
specification.” In re Morris, 127 F.3d 1048, 1054 (Fed. Cir. 1997). “[T]he
protocol of giving claims their broadest reasonable interpretation . . . does
not include giving claims a legally incorrect interpretation” “divorced from
the specification and the record evidence.” Microsoft Corp. v. Proxy conn,
Inc., 789 F.3d 1292, 1298 (Fed. Cir. 2015) (quoting In re Skvorecz, 580 F.3d
1262, 1267 (Fed. Cir. 2009) and In reNTP, Inc., 654 F.3d 1279, 1288 (Fed.
Cir. 2011)), overruled on other grounds by Aqua Products, Inc. v. Matal,
872 F.3d 1290 (Fed. Cir. 2017).
With regard to the disputed claim limitation, we agree with the
Examiner that, based on the description of modeling in paragraph 21 of the
Specification, that the recited “modeling ... to form a texture model” is
broad enough to encompass PatchMatch algorithm taught in Barnes.
Specifically, the Specification links the PatchMatch algorithm to modeling:
Texture modeling techniques are described in the
following. For example, texture modeling may be performed
for a single image. To discover self-similar patterns within the
image, a patch matching technique may be enhanced to find k-
nearest neighbor (k-NN) patches across scales (e.g., resizing),
rotations, skews, and perspective distortions. An example of a
6
Appeal 2017-008227
Application 14/092,210
patch matching technique is described as The Generalized
Patch Mutch Correspondence Algorithm, by C. Barnes, E.
Shechtman, A. Finkelstein, and D.B. Goldman, In European
Conference on Computer Vision, Sept. 2010, [that is
Barnes,] the disclosure of which is hereby incorporated by
reference in its entirety. Various other techniques are
described that involve the discovery of a texel, which is a
canonical texture element that repeats to form the texture
image.
Spec. 21 (emphasis added); see also id. ^ 59. Accordingly, we are not
persuaded that the Examiner’s conclusion concerning the scope of the claim
is erroneous. Because Appellant’s arguments are premised on a narrower
claim construction, the arguments are not commensurate with the scope of
the claims and are thus unpersuasive. See In re Self, 671 F.2d 1344, 1348
(CCPA 1982).
Appellant also argues the Examiner erred in finding that a person of
ordinary skill in the art would have combined Barnes and Shechtman. Br.
21-22. According to Appellant, “[bjecause Shechtman does not actually
add anything to Barnes, the Office’s proposed combination of Barnes and
Shechtman results in a combination yielding no actual improvement.” Id. at
22.
The Examiner finds “Barnes extends and generalizes Shechtman’s
PatchMatch algorithm, resulting in the ‘generalized PatchMatch algorithm.’”
Ans. 12-13 (quoting Br. 12). The Examiner further finds “Shechtman was
combined to further illustrate the capabilities of PatchMatch, such as the
image editing features that can be done using the algorithm before the
generalization.” Id. at 13; see also Final Act. 5 (“One of the reasons [for the
combination] would have been to find matches between image patches using
the PatchMatch algorithm.”).
7
Appeal 2017-008227
Application 14/092,210
We are not persuaded by Appellant’s arguments that the Examiner
erred. Instead, we agree with the Examiner. A person of ordinary skill in
the art would have combined the teachings of both Shechtman and Barnes as
they relate to different aspects of the PatchMatch algorithm. It is not
relevant that there may be some—or even substantial—overlap in the
teachings of the references.
Appellant further argues because Racape “does not cure the
deficiencies of Barnes and Shechtman as argued above relating to
modeling,” the Examiner did not establish a prima facie case of
unpatentability. Br. 22. However, “Racape was not used to teach the
‘modeling’ limitation in the claim,” Ans. 13, and, for the reasons discussed
above, we are not persuaded by Appellant’s arguments that there were any
deficiencies.
Accordingly, we sustain the Examiner’s rejection of claim 1, along
with the rejections of dependent claims 2-9, which were not argued
separately. See Br. 24.
Claims 11—13
Appellant argues the Examiner erred in finding the prior art teaches
the limitations recited in claim 11 for the same reasons discussed above for
claim 1. See Br. 24-26. That is, according to Appellant, because Barnes
does not teach generating a model, it cannot teach the specific modelling
recited in claim 11. See id. 27-30; cf. Ans. 13 (“Note that [the] response to
Appellant's arguments similar to ones already set forth in discussion of
independent claim 1 by Appellant are also operative in this section.”).
However, for the same reasons discussed above for claim 1, we are not
8
Appeal 2017-008227
Application 14/092,210
persuaded by Appellant’s argument that the cited prior art does not teach
modelling.
Additionally, Appellant argues the Examiner erred in finding Barnes
teaches “generating a model by the one or more computing devices which
models regularity of the located self-similar patterns,” as recited in claim 11.
See Br. 30-31. More specifically, Appellant argues:
But while Barnes does describe using the generalized
PatchMatch algorithm in conjunction with another algorithm,
this is not an example of “generating a model. . . which models
regularity of the located self-similar patterns” as recited in
claim 11. Instead, Barnes is merely describing the use of the
generalized PatchMatch algorithm in conjunction with another
algorithm for finding translational symmetries.
Id.
The Examiner finds “Bames-Shechtman-Racape teaches
image/texture manipulation using the result (model) of the PatchMatch
algorithm.” Ans. 13 (citing Barnes Figs. 1, 5-8; Shechtman Figs. 1, 4-6, 8-
12). The Examiner further finds “one of ordinary skill in the art would
recognize that fitting a lattice to analyze the portion in the user's marking in
Shechtman Fig. 9(a) reveals how the texels are repeated to [sic] for scaling
up in Fig. 9(b) while maintaining symmetry.” Id. at 13-14
We are not persuaded by Appellant’s arguments that the Examiner
erred. Instead, we determine that Appellant’s argument is premised on the
improperly narrow claim construction discussed above regarding claim 1.
To the contrary, we find the Examiner’s findings are supported by the
record.
Finally, Appellant argues Barnes does not “models regularity,” as
recited in claim 11 because Barnes “describes finding translations and not
9
Appeal 2017-008227
Application 14/092,210
transformations.” Br. 31. According to Appellant, because Barnes “does
not describe finding translations and transformations” {id.), it does not teach
the models regularity limitation.
We are not persuaded by Appellant’s arguments. First, Appellant’s
argument is not commensurate with the scope of the claim. The word
“transformations”—or “translations” for that matter—is not recited in claim
11, and Appellant does not argue persuasively why the claim should be
construed to require transformations. Second, even if such claim limitation
exists, we agree with the Examiner’s determination that skew is a type of
transformation and that Barnes Figure 6 “depicts the skewness of the
perspective of the elements.” Ans. 14.2
Accordingly, we sustain the Examiner’s rejection of claim 11, along
with the rejection of dependent claims 12 and 13, which are not separately
argued. See Br. 32.
Claims 14—20
Appellant argues “[t]he arguments presented above in regards to the
rejections of claim 1 and claim 11 apply equally to the rejection of claim
14.” Br. 33. Specifically, Appellant argues “Barnes does not describe the
generalized PatchMatch algorithm as modeling regularity of texels in image
data, and Shechtman and Racape do not cure the deficiencies of Barnes.” Id.
For the same reasons discussed above with regard to claims 1 and 11, we are
not persuaded that the Examiner erred.
2 Because we find no deficiencies associated with the Examiner’s findings
regarding Barnes, Appellant’s argument that neither Shechtman nor Racape
cures those deficiencies is moot. See Br. 32.
10
Appeal 2017-008227
Application 14/092,210
Accordingly, we sustain the Examiner’s rejection of claim 14, along
with the rejections of claims 15-20, which are not argued separately. Br. 34.
DECISION
For the above reasons, we affirm the Examiner’s decisions rejecting
claims 1-9 and 11-20.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l)(iv). See 37
C.F.R. § 41.50(f).
AFFIRMED
11