Ex Parte Chang et al

Patent Trial and Appeal Board

Sep 30, 2016

12019850 (P.T.A.B. Sep. 30, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
12/019,850 01125/2008
109673 7590 10/04/2016
McClure, Qualey & Rodack, LLP
3100 Interstate North Circle
Suite 150
Atlanta, GA 30339
FIRST NAMED INVENTOR
Te-Hao Chang
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.
251314-5170 1772
EXAMINER
RIV AS, SALVADORE
ART UNIT PAPER NUMBER
2479
NOTIFICATION DATE DELIVERY MODE
10/04/2016 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):
uspatents@mqrlaw.com
dan.mcclure@mqrlaw.com
gina.silverio@mqrlaw.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte TE-HAO CHANG, SIOU-SHEN LIN,
and CHIN-CHU AN LIANG 1
Appeal2014-009329
Application 12/019,850
Technology Center 2400
Before DEBRA K. STEPHENS, JASON V. MORGAN, and
MICHAEL J. ENGLE, Administrative Patent Judges.
MORGAN, Administrative Patent Judge.
DECISION ON APPEAL
Introduction
This is an appeal under 35 U.S.C. § 134(a) from the Examiner's final
rejection of claims 1, 7-12, 18, 19, and25-32. Claims2---6, 13-17, and20-
24 are canceled. App. Br. A-1, A-3, A-4 (Claims App'x). We have
jurisdiction under 35 U.S.C. § 6(b ).
We AFFIRM-IN-PART.
1 Appellants identify MediaTek, Inc., as the real party in interest. App. Br.
2.
Appeal2014-009329
Application 12/019,850
Invention
Appellants disclose a motion estimation method and an integrated
circuit capable of video processing. Spec. i-f 1.
Exemplary Claims
Claims 1, 8, 12, 27, and 29, reproduced below with key limitations
emphasized, are exemplary:
1. An integrated circuit capable of video processmg,
compnsmg:
an image information generator, receiving reference and
current video data to determine image information;
an adaptive motion vector (MV) candidate generator,
coupled to the image information generator, receiving the
reference and current video data to generate a set of MV
candidates, generating at least an updated MV candidate based
on the image information, and outputting the set of MV
candidates regarding to the updated MV candidate; and
a block matching unit, coupled to the adaptive MV
candidate generator, receiving the set of MV candidates and
performing block matching to determine an output MV;
wherein the image information is a motion speed
determined by motion vectors of the video data, and the
adaptive MV candidate generator generates the updated MV
candidate with reference to the motion speed.
8. The integrated circuit of claim 7, wherein the adaptive
MV candidate generator further generates a random vector,
and determines to add a predetermined difference to the
random vector, or subtract the predetermined difference from
the random vector according to the image information, to
generate the updated MV candidate.
2
Appeal2014-009329
Application 12/019,850
12. An integrated circuit capable of video processmg,
compnsmg:
an image information generator, receiving reference and
current video data to determine image information;
an adaptive motion vector (MV) candidate generator,
coupled to the image information generator, determining a
search range size based on the image information,
and receiving the reference and current video data to
generate MV candidates according to the search range size; and
a block matching unit, coupled to the adaptive MV
candidate generator, receiving the MV candidates and perform
block matching to determine an output MV therefrom;
wherein the image information is a motion speed
determined by motion vectors of the video data, and the
adaptive MV candidate generator adjusts the search range size
with reference to the motion speed.
27. An integrated circuit capable of video processmg,
compnsmg:
an image information generator, receiving reference and
current video data to determine image information;
an adaptive motion vector (MV) candidate generator,
coupled to the image information generator, receiving the
reference and current video data to generate a set of MV
candidates, generating at least an updated MV candidate based
on the image information, and outputting the set of M V
candidates regarding to the updated MV candidate; and
a block matching unit, coupled to the adaptive MV
candidate generator, receiving the set of MV candidates and
performing block matching to determine an output MV;
wherein the image information is a global motion speed
determined by an average of MV s at predetermined locations of
the video data, and the adaptive MV candidate generator
generates the updated MV candidate with reference to the
global motion speed.
3
Appeal2014-009329
Application 12/019,850
29. An integrated circuit capable of video processmg,
compnsmg:
an image information generator, receiving reference and
current video data to determine image information;
an adaptive motion vector (MV) candidate generator,
coupled to the image information generator, receiving the
reference and current video data to generate a set of MV
candidates, generating at least an updated MV candidate based
on the image information, and outputting the set of MV
candidates regarding to the updated MV candidate; and
a block matching unit, coupled to the adaptive MV
candidate generator, receiving the set of MV candidates and
performing block matching to determine an output MV;
the image information is a block difference determined
by a sum of differences between pixels of blocks of the reference
and current video data, and the adaptive MV candidate
generator generates the updated MV candidate with reference
to the block difference.
Rejections
The Examiner rejects claims 1, 10, 11, and 19 under 35 U.S.C.
§ 103(a) as being unpatentable over Ye et al. (US 2006/0133495 Al,
published June 22, 2006) (hereinafter "Ye"), Tappermann et al. (US
2010/0165210 Al, published July 1, 2010) (hereinafter "Tappermann"), and
Hirano et al. (US 6,509,930 Bl, issued Jan. 21, 2003) (hereinafter "Hirano").
Final Act. 3-11.
The Examiner rejects claim 7 under 35 U.S.C. § 103(a) as being
unpatentable over Ye, Tappermann, Hirano, and Handjojo et al. (US 2002/
0171759 Al, published Nov. 21, 2002) (hereinafter "Handjojo"). Final Act.
11-12.
4
Appeal2014-009329
Application 12/019,850
The Examiner rejects claims 8, 9, 25, and 26 under 35 U.S.C. § 103(a)
as being unpatentable over Ye, Tappermann, Hirano, Handjojo, and Bellers
(US 2004/0071215 Al, published Apr. 15, 2004). Final Act. 12-14.
The Examiner rejects claim 12 under 35 U.S.C. § 103(a) as being
unpatentable over Tappermann, Kawashima (US 2006/0215758 Al,
published Sept. 28, 2006), and Hirano. Final Act. 15-18.
The Examiner rejects claim 18 under 35 U.S.C. § 103(a) as being
unpatentable over Ye, Tappermann, Kawashima, and Kang (US 2004/
0114688 Al, published June 17, 2004). Final Act. 18-19.
The Examiner rejects claims 27 and 28 under 35 U.S.C. § 103(a) as
being unpatentable over Ye, Tappermann, and Edanami (US 2003/0076997
Al, published Apr. 24, 2003). Final Act. 19-25.
The Examiner rejects claims 29 and 30 under 35 U.S.C. § 103(a) as
being unpatentable over Ye, Tappermann, and Gharavi (US 5,200,820,
issued Apr. 6, 1993). Final Act. 25-31.
The Examiner rejects claim 31 under 35 U.S.C. § 103(a) as being
unpatentable over Tappermann, Kawashima, and Gharavi. Final Act. 31-34.
The Examiner rejects claim 32 under 35 U.S.C. § 103(a) as being
unpatentable over Tappermann, Kawashima, and Edanami. Final Act. 35-
38.
ISSUES
1. Did the Examiner err in combining the teachings and suggestions
of Ye, Tappermann, and Hirano to show that the claim 1 recitation "the
adaptive MV candidate generator generates the updated MV candidate with
reference to the motion speed" would have been obvious?
5
Appeal2014-009329
Application 12/019,850
2. Did the Examiner err in combining the teachings and suggestions
of Ye, Tappermann, Hirano, and Handjojo in the manner of claim 7?
3. Did the Examiner err in finding the combination of Ye,
Tappermann, Hirano, Handjojo, and Bellers teaches or suggests "wherein
the adaptive MV candidate generator further generates a random vector, and
determines to add a predetermined difference to the random vector, or
subtract the predetermined difference from the random vector according to
the image information, to generate the updated MV candidate," as recited in
claim 8?
4. Did the Examiner err in finding the combination of Tappermann,
Kawashima, and Hirano teaches or suggests "the adaptive MV candidate
generator adjusts the search range size with reference to the motion speed,"
as recited in claim 12?
5. Did the Examiner err in finding the combination of Ye,
Tappermann, and Edanami teaches or suggests "the adaptive MV candidate
generator generates the updated MV candidate with reference to the global
motion speed," as recited in claim 27?
6. Did the Examiner err in finding the combination of Ye,
Tappermann, and Gharavi teaches or suggests "the adaptive MV candidate
generator generates the updated MV candidate with reference to the block
difference," as recited in claim 29?
ANALYSIS
Claims 1, 10, 11, and 19
In rejecting claim 1, the Examiner finds Tappermann's method for
motion estimation, which generates updated predictor vectors (UPV) based
on predictor vectors (PV), and applies the UPV s to the PV s to generate a
6
Appeal2014-009329
Application 12/019,850
limited set of candidate vectors (CV), teaches or suggests an adaptive
motion vector (MV) candidate generator that generates an updated MV
candidate with reference to image information. Final Act. 4--5 (citing
Tappermann Figs. 2--4, i-f 16). The Examiner finds that Hirano's motion
speed and direction detector teaches or suggests the image information being
motion speed. Final Act. 5---6 (citing Hirano Figs. 13A, 14A, col. 7, 11. 62-
64).
Appellants contend the Examiner erred because "the updated UPV in
Tappermann is an updated one for the predictor vector PV, instead of for the
motion v[ e ]ctor candidate." App. Br. 11. Appellants argue that "the
predictor vector PV in Tappermann . .. is an element participat[ing] in
producing the motion v[ e ]ctor candidate, instead of being the motion
v[ e ]ctor candidate itself." Id. However, Appellants' arguments do not
persuasively distinguish the claimed motion vector candidate from
Tappermann's updated predictor vector. See Ans. 2 (citing Tappermann
Figs. 2--4, i-f 16). Appellants' arguments also do not persuasively distinguish
the claimed updated motion vector candidate from a candidate vector in
Tappermann' s limited set of candidate vectors. See Ans. 2 (citing
Tappermann Figs. 2--4, i-f 16). Therefore, we agree with the Examiner that
Tappermann teaches or suggests an "adaptive MV candidate generator
generates the updated MV candidate," as recited in claim 1.
Appellants further argue that "although Hirano . .. makes use[] of the
motion speed ... , the motion speed is used for selecting an interpolation
signal, NOT the MV candidates for motion estimation." App. Br. 12.
However, the Examiner merely relies on the type of information that can be
used as image information. See Ans. 3 (citing Hirano Figs. 13A, 14A, col. 7,
7
Appeal2014-009329
Application 12/019,850
11. 62---64). Appellants argue that the "interpolation signal and the MV
candidates for the motion estimation are two distinctly different features in
video decoding." App. Br. 12. However, the Examiner correctly finds that
"interpolation could not take place without motion vectors." Ans. 4. Thus,
Appellants' arguments do not show error in the Examiner's use of Hirano's
motion speed to modify Tappermann's updated motion vector candidate
generation. See Final Act. 5---6.
Appellants further argue that "the tenuous combination of Ye,
Tappermann and Hirano is improper and therefore does not render the
claims obvious." App. Br. 14. In particular, Appellants argue the Examiner
does not explain an "objective rationale ... that would have motivated the
combination." Id. However, the Examiner does not need to provide a
motivation to show obviousness. Rather, there merely must be "some
articulated reasoning with some rational underpinning to support the legal
conclusion of obviousness." KSR Int 'l Co. v. Teleflex Inc., 550 U.S. 398,
418 (2007) (quoting In re Kahn, 441F.3d977, 988 (2006)). The Examiner
has presented articulated reasoning having a rational underpinning to show
the proposed combination would have been obvious. Final Act. 5-6; Ans.
3-6. Appellants' arguments do not persuasively show error in the
Examiner's articulated reasoning. Therefore, we do not find error in the
Examiner's reliance on the combined teachings and suggestions of Ye,
Tappermann, and Hirano.
For these reasons, we agree with the Examiner that it would have been
obvious to combine Ye, Tappermann, and Hirano to teach or suggest "the
adaptive MV candidate generator generates the updated MV candidate with
reference to the motion speed," as recited in claim 1. Accordingly, we
8
Appeal2014-009329
Application 12/019,850
sustain the Examiner's 35 U.S.C. § 103(a) rejection of claim 1, and claims
10, 11, and 19, which Appellants do not argue separately. App. Br. 16.
Claim 7
With respect to claim 7, Appellants argue that Handjojo fails to cure
the alleged deficiencies of Ye, Tappermann, and Hirano. App. Br. 17.
However, as we do not find Ye, Tappermann, and Hirano to be deficient as
alleged, we do not find Appellants' arguments persuasive.
Appellants also raise an argument with respect to Bellers in
contending the Examiner erred in rejecting claim 7. Id. However, the
Examiner does not rely on Bellers in rejecting claim 7. Final Act. 11-12.
Therefore, Appellants' argument regarding Bellers is not applicable with
respect to the Examiner's rejection of claim 7.
Finally, Appellants contend the Examiner did not provide "objective
or evidentiary support" for combining the teachings and suggestions of Ye,
Tannermann. Hirano. and Handioio. Ann. Br. 17. However. Annellants' _._ _._ / / .J .J _._ _._ / _._ _._
conclusory allegation fails to show error in the Examiner's articulated
reasoning that the use of Handjojo in the manner claimed would have
reduced motion vector complexity. See Final Act. 12; Ans. 7. Therefore,
Appellants' arguments do not persuasively show the Examiner erred in
combining the teachings and suggestions of Ye, Tappermann, Hirano, and
Handjojo in the manner of claim 7.
Accordingly, we sustain the Examiner's 35 U.S.C. § 103(a) rejection
of claim 7.
Claims 8, 9, 25, and 26
In rejecting claim 8, the Examiner finds that the generation of a
random vector in Bellers teaches or suggests wherein the adaptive MV
9
Appeal2014-009329
Application 12/019,850
candidate generator further generates a random vector, and determines to
add a predetermined difference to the random vector, or subtract the
predetermined difference from the random vector according to the image
information, to generate the updated MV candidate. Final Act. 13 (citing
Bellers i-fi-12, 13). Appellants contend the Examiner erred because Bellers
merely "describes a random vector generator ... but reveals no further data
processing to the random vector, including the addition, subtraction,
multiplication, or division" of claims 8 and 9. App. Br. 18.
In response to Appellants' argument, the Examiner find that the
execution of a sum-of-absolute differences (SAD) method or a mean-
squared-error (MSE) method in Bellers teaches or suggests the claimed
mathematical operations. Ans. 9 (citing Bellers i-f 16). However, the cited
portion of Bellers relates to the comparison of picture blocks of one picture
with multiple picture blocks in another picture. Bellers i-f 16. Although
SAD and MSE are used for evaluation of error given a particular vector, we
determine addition, subtraction, multiplication, or division is not performed
on a vector as part of the two disclosed methods in Bellers. Id. Thus, the
Examiner's findings do not show that Bellers teaches or suggests "wherein
the adaptive MV candidate generator further generates a random vector, and
determines to add a predetermined difference to the random vector, or
subtract the predetermined difference from the random vector according to
the image information, to generate the updated MV candidate," as recited in
claim 8.
The Examiner does not show that Ye, Tappermann, Hirano, or
Handjojo cure the noted deficiency of Bellers. Accordingly, we do not
sustain the Examiner's 35 U.S.C. § 103(a) rejection of claim 8. Claims 9,
10
Appeal2014-009329
Application 12/019,850
25, and 26 are similar to claim 8, although claims 9 and 26 recites
multiplying or dividing a randomly generated vector by a predetermined
gain, rather than adding or subtracting. The Examiner's findings are
similarly deficient with respect to these claims. Therefore we also do not
sustain the Examiner's 35 U.S.C. § 103(a) rejection of claim 9, 25, and 26.
Claims 12, 18, 31, and 32
In rejecting claim 12, the Examiner finds Kawashima's setting of a
search range during motion compensation teaches or suggests the adaptive
MV candidate generator adjusts the search range size with reference to the
motion speed. Final Act. 16 (citing Kawashima i-f 9). Appellants contend
the Examiner erred because in Kawashima "the rectangular search range is
merely afixed range around the search center." App. Br. 20 (emphasis
added). Appellants argue "Kawashima does not propose to change the size
of the rectangular shape, nor does Kawashima disclose to determine a search
ram!e size for the search ram!e." Id.
~ ~
The Examiner finds Kawashima teaches the motion-vector-search unit
sets the search range based on information related to a motion vector
detected. Ans. 13 (citing Kawashima i-f 9). However, we find that although
Kawashima sets (i.e., moves) the search range with reference to a motion
vector detected in the search range (Kawashima i-f 9), the Examiner has not
shown nor explained why Kawashima teaches, or at least suggests, adjusting
the search range size. The Examiner also finds that interpolation could not
occur without motion vectors, pointing to Hirano' s Motion Speed and
Direction Detector (Ans. 13-14). The Examiner has not sufficiently
explained how the combination would have taught of suggested the disputed
limitation.
11
Appeal2014-009329
Application 12/019,850
Further, the Examiner finds that adjusting the search range size is not
"patentably significant since it at most relates to size of the article under
consideration which is not ordinarily a matter of invention." Ans. 13 (citing
In re Yount, 171F.2d317 (CCPA 1948); In re Rose, 220 F.2d 459 (CCPA
1955); MPEP 2144.04(IV)(A)). However, the disputed claim recitation is
not directed to "mere size," Yount, 171 F .2d at 318, such as scaling a prior
art invention to be bigger or smaller without changing the device to
"perform[] and operate[] any differently from the prior art." Gardner v.
TEC Sys., Inc., 725 F.2d 1338, 1349 (Fed. Cir. 1984). Rather, the claim is
directed to adjusting the search range size relative to the motion speed, since
motion speed may change the expected variance within a frame. See Spec.
i-f 29. Such a change does cause the device to perform and operate
"differently from the prior art," Gardner, 725 F.2d at 1349, thus making the
Examiner's case law and MPEP citations inapposite to the facts here. We
conclude that reading out the adjusts the search range size with reference to
the motion speed recitation from claim 12 is unreasonable. Thus, we
determine the Examiner's findings do not show that the combination of
Tappermann, Kawashima, and Hirano teaches or suggests "the adaptive MV
candidate generator adjusts the search range size with reference to the
motion speed," as recited in claim 12.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 103(a)
rejection of claim 12. The Examiner does not show that Ye, Kang, Gharavi,
or Edanami cure the noted deficiency with respect to similar recitations in
claims 18, 31, and 32. Therefore, we also do not sustain the Examiner's 35
U.S.C. § 103(a) rejections of these claims.
12
Appeal2014-009329
Application 12/019,850
Claims 27 and 28
In rejecting claim 27, the Examiner finds Tappermann's method for
motion estimation teaches or suggests the adaptive MV candidate generator
generates the updated MV candidate with reference to digital image data.
Final Act. 21 (citing Tappermann Figs. 2--4, i-f 16). The Examiner further
finds that Edanami' s calculation of an average value of motion amounts
teaches or suggests the digital image data being global motion speed
determined by an average of MVs at predetermined locations of the video
data. Final Act. 21-22 (citing Edanami i-fi-110, 84).
Appellants' arguments with respect to the Examiner's reliance on
Tappermann to reject claim 27 are similar to those made with respect to the
Examiner's reliance on the same teachings and suggestions to reject claim 1.
App. Br. 23-24. For the same reasons as discussed above, we do not find
Appellants' arguments persuasive of error.
Appellants further argue that although Edanami mentions an average
moving speed vector, it "still fails to disclose generating the updated MV
candidate with reference to the global motion speed." App. Br. 25.
However, the Examiner merely relies on Edanami to teach or suggest that
global motion speed determined in the claimed manner is a form of digital
image data. Final Act. 21-22. The Examiner relies on Tappermann, not
Edanami, for the use of such information. Furthermore, Appellants'
arguments do not persuasively distinguish the claimed global motion speed,
and its determination, from Edanami' s average value of motion amounts.
Appellants further argue that the Examiner provided "no objective
rationale ... that would have motivated the combination" of Ye,
Tappermann, and Edanami. App. Br. 26. Appellants' arguments are similar
13
Appeal2014-009329
Application 12/019,850
to the arguments they make with respect to the comb inability of Ye,
Tappermann, and Hirano. Id. at 14. For similar reasons as those discussed
above, we do not find Appellants' arguments regarding the comb inability of
Ye, Tappermann, and Edanami persuasive.
For these reasons, we agree with the Examiner that the combination of
Ye, Tappermann, and Edanami teaches or suggests "the adaptive MV
candidate generator generates the updated MV candidate with reference to
the global motion speed," as recited in claim 27. Accordingly, we sustain
the Examiner's 35 U.S.C. § 103(a) rejection of claim 27, and of claim 28,
which Appellants do not argue separately. App. Br. 26.
Claims 29--31
In rejecting claim 29, the Examiner finds Gharavi's consideration of
frame-to-frame displacement of moving objects in a coding process, which
includes storing coded differences between the intensities of pels in the
block of a current frame and the intensities of pels in a previous frame,
teaches or suggests the adaptive MV candidate generator generates the
updated MV candidate with reference to the block difference, which is
determined by a sum of differences between pixels of blocks of the reference
and current video data., Final Act. 28 (citing Gharavi col. 1, 11. 7-10, col. 7,
11. 26-31).
Appellants contend the Examiner erred because "the block difference
in claim 29 is determined by a sum of differences between pixels of blocks
of the reference and current video data." App. Br. 28. Appellants argue that
"Gharavi only mentions coded differences between the intensities of the pels
in the block in the current frame and the intensities of the pels in the
14
Appeal2014-009329
Application 12/019,850
particular block in the previous frame, instead of a sum of the coded
differences." Id.
The Examiner finds "Gharavi teaches the image information (read as
picture elements) is a block difference determined by differences between
pixels of blocks of the reference and current video data." Ans. 21 (citing
Gharavi col. 7, 11. 26-31 ). However, the Examiner has not shown that the
block difference in Gharavi teaches or suggests a sum of differences between
pixels. Moreover, the Examiner does not show that Ye or Tappermann cure
the noted deficiency of Gharavi. Therefore, we agree with Appellants that
the Examiner's findings do not show the combination of Ye, Tappermann,
and Gharavi teaches or suggests "the adaptive MV candidate generator
generates the updated MV candidate with reference to the block difference,"
as recited in claim 29.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 103(a)
rejection of claim 29, and claim 30, which contains a similar recitation and is
similarly rejected. The Examiner also does not show that Kawashima cures
the noted deficiency of Gharavi. Therefore, for these same reasons-
independent of the reasons discussed above based on recitations in claim 31
that are similar to disputed recitations in claim 12-we also do not sustain
the Examiner's 35 U.S.C. § 103(a) rejection of claim 31.
15
Appeal2014-009329
Application 12/019,850
DECISION
We affirm the Examiner's decision rejecting claims 1, 7, 10, 11, 19,
27, and 28.
We reverse the Examiner's decision rejecting claims 8, 9, 12, 18, 25,
26, and 29-32.
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.
§ 41.50(±).
AFFIRMED-IN-PART
16