Ex Parte JAIN et al

Patent Trial and Appeal Board

May 26, 2016

12577157 (P.T.A.B. May. 26, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
12/577, 157 10/10/2009 CHIRAGJAIN
40317 7590 05/31/2016
GLOBAL IP SERVICES, PLLC
10 CRESTWOOD LANE
NASHUA, NH 03062
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
ATTORNEY DOCKET NO. CONFIRMATION NO.
l 864.054US 1 3643
EXAMINER
CARTER, RICHARD BRUCE
ART UNIT PAPER NUMBER
2485
NOTIFICATION DATE DELIVERY MODE
05/31/2016 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):
pnama@globalipservices.com
docketing@globalipservices.com
pradeep@globalipservices.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte CHIRAG JAIN and SRIRAM SETHURAMAN
Appeal2014-002168
Application 12/577,157
Technology Center 2400
Before JASON V. MORGAN, MELISSA A. RAAP ALA, and
NABEEL U. KHAN, Administrative Patent Judges.
RAAP ALA, Administrative Patent Judge.
DECISION ON APPEAL
This is a decision on appeal under 35 U.S.C. § 134(a) from a final
rejection of claims 1-35. We have jurisdiction under 35 U.S.C. § 6(b ). We
reverse.
Appeal2014-002168
Application 12/577,157
Il~VENTION
Appellants' invention is "a system and method for robust
spatiotemporal combining for video enhancement." Spec. i-f 1. Claim 16,
reproduced below, is illustrative of the subject matter on appeal.
16. A spatiotemporal combining method for video
enhancement, comprising:
computing a standard deviation estimate between a video frame
and a temporally neighboring frame of the video frame in the
video sequence;
computing an error value between a pixel value in a current video
frame and corresponding motion compensated pixel value in a
temporally neighboring video frame of the current video frame;
computing a temporal weighting factor for each pixel as a
function of the error value and the standard deviation estimate;
and
combining the pixel value in the current video frame and its
corresponding motion compensated pixel value in the temporally
neighboring video frame using the computed temporal weighting
factor.
REJECTIONS ON APPEAL
Claim 16 stands rejected under 35 U.S.C. § 102(e) as being
anticipated by Kameyama (US 2011/0255610 Al; Oct. 20, 2011). Final Act.
6-7.
Claim 25 stands rejected under 35 U.S.C. § 102(e) as being
anticipated by Tian (US 2008/0151101 Al; June 26, 2008). Final Act. 7-8.
Claims 1, 2, 5, 6, 10, 11, 12, 13, 15, 19, 28, 30, and 31 stand rejected
under 35 U.S.C. § 103(a) as being obvious over the combination of
2
Appeal2014-002168
Application 12/577,157
Kameyama and Huang (US 2010/0290530 Al; Nov. 18, 2010). Final Act.
9-18.
Claims 21, 23, 27, and 32-35 stand rejected under 35 U.S.C. § 103(a)
as being obvious over the combination of Tian and Huang. Final Act.
18-27.
Claims 3, 4, 7-9, 14, 17, 18, 20, 22, 24, 26, and 29 stand rejected
under 35 U.S.C. § 103(a) as being obvious over the combination of
Kameyama, Huang, and Tian. Final Act. 27-34.
ISSUES
Appellants' contentions present the following issues: 1
(1) Did the Examiner establish that Kameyama discloses
computing a weighting factor as a function of an error value and standard
deviation estimate, as required by independent claims 1, 11, and 16?
(2) Did the Examiner establish that Tian discloses computing a
weighting factor as a function of an error value and standard deviation
estimate, as required by independent claims 21, 23, 25, 27, and 32?
ANALYSIS
As reflected by the above issue statements, each of the independent
claims is rejected over Kameyama or Tian (alone or in combination with
other references) and requires computing a weighting factor as a function of
an error value and standard deviation estimate. More particularly, a
weighting factor is computed for a singular image element of given
1 We do not reach the additional contentions presented by Appellants
because the identified issues are dispositive of the appeal.
3
Appeal2014-002168
Application 12/577,157
granularity as a function of both an error value determined for the same
granularity and a standard deviation estimate determined for a different
granularity. See e.g., supra, claim 16 ("weighting factor for each pixel,"
"error value between a pixel value ... and ... [another] pixel value," and
"standard deviation estimate between a video frame and [another] frame") ..
Thus, for whatever prior art element is mapped by a rejection to an
independent claim's weighting factor, the Examiner's findings must
establish that a singular such prior art element is computed from both an
error value and a standard deviation estimate; i.e., establish that a singular
computation of such a prior art element inputs both an error value and
standard deviation estimate.
Each of independent claims 1, 11, and 16 is rejected over Kameyama
alone or in combination with other references. For these claims, the
Examiner maps the weighting factor to Kameyama's correlation value, the
error value to Kameyama's mean square error of Formula 1, and the
standard deviation estimate to Kameyama' s standard-deviation difference
value of the histograms for the coordinate-transformed frame Frru and
reference frame FrN. Ans. 6, 10.2 Appellants contend Kameyama teaches
2 The Examiner reads claim 16' s at-issue claim elements on Kameyama as
follows: reads the weighting factor on paragraphs 290, 323 and thereby on
Kameyama's correlation value dO(x, y); reads the error value on paragraphs
228, 290 and thereby on Kameyama's mean square error; and reads the
standard deviation estimate on paragraph 323 and thereby on Kameyama's
standard-deviation difference value. Final Act. 6-7. For claims 1 and 11,
the Examiner states Huang's paragraphs 39, 85, 152, and 187-88 teach the
at-issue computation of a weighting factor. Final Act. 10, 16; Ans. 7. The
Examiner relies upon the cited teachings of Huang as merely suggesting to
implement Kameyama's above elements at the "sub-block" granularity
required by claims 1 and 11. See Huang ,-r,-r 83-85 (sum of absolute
4
Appeal2014-002168
Application 12/577,157
the cited correlation value as computed from either the mean square error or
standard-deviation difference value; not as computed from both. Br. 12.
The argument is persuasive insofar as Kameyama's cited sections do not
expressly teach a singular correlation value that is a function of both the
mean square error and standard-deviation difference value. Rather, the cited
sections teach the "correlation value can be computed as a mean square
error" by a "correspondent relationship estimation means 2" (Kameyama i-fi-1
289-90) and a "standard-deviation difference value ... may be employed as
the correlation value" by a "correlation-value computation means 6" (id. at
i-fi-1322-23). See also id. at i1335 (relating the "correspondent relationship
estimation means 2" and "correlation-value computation means 6"). That is,
the cited sections teach a correlation value that is a function of a mean
square error or standard-deviation difference value. Thus, the Examiner has
not established Kameyama discloses a weighting factor that is a function of
both the error value and the standard deviation estimate.
Each of independent claims 21, 23, 25, 27, and 32 is rejected over
Tian alone or in combination with other references. For these claims, the
Examiner maps the weighting factor to Tian's "weighted median filter
(Wmed)," the error value to Tian's "summed absolute differences (SAD),"
and the standard deviation estimate to Tian's "sum of squared errors (SSE)."
Ans. 8-9, 11-12. 3 Appellants contend Tian's cited sections describe using
difference (SAD) values for sub-blocks), 152 (weighting factors for
sub-blocks). Moreover, we did not observe a singular weighting factor
computed from both an error value and a standard deviation estimate.
3 The Examiner reads claim 25 on Tian as follows: reading the weighting
factor on paragraphs 159, 190, and 209-10 and thereby on Tian's weighted
median filter (Wmed) value; reading the error value on paragraphs 190 and
5
Appeal2014-002168
Application 12/577,157
sum of squared errors (SSE) to measure the similarity between a predicting
block and a predicted block for the Wmed current frame 2060 relative to the
Wmed next frame 2058 and determining SAD/SSD for every block in
forward and backward difference metric associated with the current frame,
but do not mention computing a temporal weighting factor as a function of
the error value and the standard deviation estimate. Br. 16. The argument is
persuasive insofar as Tian's cited sections do not disclose a singular Wmed
value that is a function of both the SAD and SSE. Rather, the cited sections
teach the SAD as a difference metric (Tian i-f 190) and the SSE as used to
measure the similarity between a predicting block for the Wmed current
frame and a predicted block for the Wmed next frame (Tian i-f 159). That is,
the cited sections teach a difference metric and Wmed information that are,
respectively, functions of the SAD and SSE; not a singular weighting factor
value that is a function of both. Thus, the Examiner has not established Tian
discloses a weighting factor that is a function of both the error value and the
standard deviation estimate.
209-210 and thereby on Tian's summed absolute differences (SAD);
reading the standard deviation estimate on paragraphs 85 and 190 and
thereby on Tian's sum of squared errors (SSE). Final Act. 7-8. For claims
21, 23, 27, and 32, the Examiner again states Huang's paragraphs 39, 85,
152, and 187-88 teach computation of a weighting factor. Id. at 10 (claim
1 ), 16 (claim 11 ); Ans. 7. Thus, the Examiner relies upon the cited teachings
as merely suggesting to implement Tian's above elements at the "sub block"
granularity required by claims 21, 23, 27, and 32. See Huang i-fi-183-85
(SAD values for sub blocks), 152 (weighting factors for sub blocks).
Moreover, we did not observe a singular weighting factor computed from
both an error value and a standard deviation estimate.
6
Appeal2014-002168
Application 12/577,157
CONCLUSION
The Examiner has not established Kameyama or Tian discloses the
independent claims' computing of a weighting factor as a function of an
error value and standard deviation estimate.
The Examiner does not find the additional Huang reference teaches
such computing of a weighting factor.
Accordingly, we do not sustain the: (i) anticipation rejection of
independent claim 16 over Kameyama; (ii) anticipation rejection of
independent claim 25 over Tian; (iii) obviousness rejection of independent
claims 1 and 11, and dependent claims 2, 5, 6, 10, 12, 13, 15, 19, 28, 30, and
31, over Kameyama and Huang; (iv) obviousness rejection of independent
claims 21, 23, 27, and 32, and dependent claims 33-35, over Tian and
Huang; and (v) obviousness rejection of dependent claims 3, 4, 7-9, 14, 17,
18, 20, 22, 24, 26, and 29 over Kameyama, Huang, and Tian.
DECISION
The Examiner's decision rejecting claims 1-35 is reversed.
REVERSED
7