Ex Parte Demos

Patent Trial and Appeal Board

Aug 2, 2016

13197211 (P.T.A.B. Aug. 2, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
13/197,211 08/03/2011
26171 7590 08/04/2016
FISH & RICHARDSON P,C (DC)
P.O. BOX 1022
MINNEAPOLIS, MN 55440-1022
FIRST NAMED INVENTOR
Gary A. Demos
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.
07314-0013004 5313
EXAMINER
LEE, YYOUNG
ART UNIT PAPER NUMBER
2485
NOTIFICATION DATE DELIVERY MODE
08/04/2016 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):
P ATDOCTC@fr.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte GARY A. DEMOS
Appeal2015-001244
Application 13/197 ,211
Technology Center 2400
Before JOHN P. PINKERTON, NATHAN A. ENGELS,
and NORMAN H. BEAMER, Administrative Patent Judges.
PINKERTON, Administrative Patent Judge.
DECISION ON APPEAL
Appellant 1 files this appeal under 35 U.S.C. § 134(a) from the
Examiner's Final Rejection of claims 1-3 and 6-18, which constitute all of
the claims pending in this application. Claims 4 and 5 are canceled. We
have jurisdiction under 35 U.S.C. § 6(b ).
We reverse.
1 Appellant identifies Dolby Laboratories Licensing Corporation as the real
party in interest. App. Br. 1.
Appeal2015-001244
Application 13/197 ,211
STATEMENT OF THE CASE
Introduction
Appellant's disclosed and claimed invention is generally directed to
improving compressed image chroma information in MPEG-like video
compression systems. Spec. 2. 2
Claim 1 is illustrative and reproduced below (with the disputed
limitation emphasized):
1. A video image system using quantization parameters of a
color video image, the system including:
a data storage system;
a decoder configured to interact with the data storage
system to execute instructions to:
receive a bias value at a decoder as part of a
compressed bit stream;
utilize a variable quantization step size and a
quantization parameter (QP) to represent a size of a step
where an increase in the QP corresponds to a larger
quantization step size;
utilize a first QP value for a Y luminance channel
of the color video image for decoding a first macro block;
utilize a second QP value for at least one of U and
V color channels of the color video image for decoding
said first macroblock, wherein said second QP value is
2 Our Decision refers to the Final Office Action mailed Mar. 6, 2014 ("Final
Act."); Appellant's Appeal Brief filed Aug. 6, 2014 ("App. Br."); the
Examiner's Answer mailed Sept. 9, 2014 ("Ans."); Appellant's Reply Brief
filed Oct. 29, 2014 ("Reply Br."); and, the original Specification filed Aug.
3, 2011 ("Spec.").
2
Appeal2015-001244
Application 13/197 ,211
determined by at least applying the bias value to the first
QP value to provide a first intermediate result.
Rejection on Appeal
Claims 1-3 and 6-18 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Naimpally et al. (US 5,294,974; issued Mar. 15, 1994)
("Naimpally"), Ribas-Corbera (US 6,535,251 Bl; issued Mar. 18, 2003), and
Haskell et al. (US 2006/0002467 Al; published Jan. 5, 2006) ("Haskell").
Issue
Appellant's arguments in the Briefs present us with the following
dispositive issue: 3
Does the combination ofNaimpally, Ribas-Corbera, and Haskell teach
or suggest the limitation
utilize a second QP value for at least one of U and V color
channels of the color video image for decoding said first
macroblock, wherein said second QP value is determined by at
least applying the bias value to the first QP value to provide a
first intermediate result;
as recited in claim 1, and as similarly recited in claims 6 and 12?
ANALYSIS
Appellant argues that
none of Naimpally, Ribas-Corbera, and Haskell, alone or
combined, discloses determining a second QP value for a color
channel for decoding a first macroblock "by at least applying the
bias value to the first QP value to provide a first intermediate
result," where the first QP value is utilized "for a Y luminance
3 Appellant's arguments raise additional issues, but we do not reach them
because the identified issue is dispositive of the appeal.
3
Appeal2015-001244
Application 13/197 ,211
channel of the color video image for decoding [the same] first
macro block."
App. Br. 9--10. Regarding the Examiner's finding that Ribas-Corbera
teaches the concept of "quantization adjustment wherein the adjusted QP
value (e.g. new QT2) is determined by at least applying a bias value (e.g. -1)
to a first QP value (e.g. QT2) to provide a first intermediate result (e.g. QT2 -
1 )" (see Ans. 3), Appellant argues Ribas-Corbera discloses a minimum
quantization step size threshold QT2 for updating the minimum quantization
step size QMIN at the encoder, "which is different from determining 'a second
QP value for at least one of U and V color channels of the color video image
for decoding said first macro block' at the decoder, as recited in claiml ."
App. Br. at 11. Regarding the Examiner's finding that Haskell teaches the
well known concept of transmitting the first QP value and the bias value as
part of a compressed bit stream, Appellant argues "while Haskell discloses
transmitting a change in Qp value in the output bitstream, the change in Qp
value is for decoding a different macroblock." Id. at 12 (citing Haskell i-fi-1
27, 42).
The Examiner finds that Ribas-Corbera and Haskell are consistent
with Appellant's Specification "in its broadest sense where a macroblock
may just be a large block comprising several small image block regions"-
"a large block of 16 x 16 pixel data that is sub-divided into four small macro
blocks of size 8 x 8 image regions." Ans. 5-6 (citing Spec. i16, Ribas-
Corbera 2:21-23, Haskell i-f 12). The Examiner also finds Haskell teaches
"receiving at the decoder 240 a first Qp value for a first 8 x 8 macroblock
and three different update bias Qp values for the other three 8 x 8
macroblocks in order to decode a single 16 x 16 macro block." Id. at 6. The
4
Appeal2015-001244
Application 13/197 ,211
Examiner further finds Ribas-Corbera teaches the value of the step size may
be modified for each block instead of for one larger macro block. Id.
We are persuaded by Appellant's arguments that the Examiner has
erred. Appellant argues, and we agree,
the broadest reasonable interpretation of the term macroblock
cannot encompass both a large block of pixels and a portion of
that same block of pixels because one of ordinary skill in the art
would not interpret a macroblock in this way .... A macroblock
may be composed of smaller blocks of pixels, but is not itself
composed of other macroblocks.
Reply Br. 2. Appellant also argues, and we agree, the Examiner's finding
that Haskell teaches "a first Qp value for a first 8 x 8 macroblock and three
different update bias Qp values for the other three 8 x 8 macroblocks" is
inconsistent with Haskell's disclosure of receiving at most a single Qp value
for a given macroblock. App. Br. 8-9 (citing Haskell i-f 27 ("Each macro
block contains as many as four blocks of luminance data and two blocks of
chrominance data. A single Qp value is used for ihe macro block. Qp is
updated on a macro block by macro block basis.")). Thus, we further agree
with Appellant's argument that:
Haskell does not disclose determining a second QP value for at
least one of U and V color channels for decoding a first
macroblock "by at least applying the bias value to the first QP
value to provide a first intermediate result," where the first QP
value is utilized "for a Y luminance channel of the color video
image for decoding [the same] first macro block."
Reply Br. 3.
Regarding the Examiner's finding that "Ribas-Corbera (i.e. col. 2 :21-
23) teaches that the value of the step size may be modified for each block
5
Appeal2015-001244
Application 13/197 ,211
instead of for one larger macro block" (see Ans. 6), we agree with
Appellant's argument:
But the cited portion of Ribas-Corbera discloses no such feature.
Column 2, lines 21 to 23 of Ribas-Corbera discloses that "the
value of the step size is often modified at each image block of a
frame in order to further guarantee that the number of bits
produced is close to the desired target." (Emphasis added) Ribas-
Corbera discloses that each image block of a frame may be
modified, not "each block instead of for one larger macro block,"
as contended by the Examiner.
Reply Br. 3.
In regard to the Examiner's finding that both Ribas-Corbera and
Haskell teach the well know concept of "updat[ing] signals by merely
adding or subtracting a number from a previous Qp value" (see Ans. 6-7),
we agree with Appellant's argument that:
Claim 1, on the other hand, recites ''utilize a first QP value for a
Y luminance channel of the color video image for decoding a
first macroblock," and "utilize a second QP value for at least one
ofU and V color channels of the color video image/or decoding
[the same] first macroblock, where the "second QP value is
determined by at least applying the bias value to the first QP
value to provide a first intermediate result." Since neither
Ribas-Corbera nor Haskell discloses applying the bias value to a
first Qp value to determine a second QP value for the same
macroblock, the rejection should be withdrawn.
Reply Br. 3--4.
Accordingly, we agree with Appellant the Examiner erred ( 1) in
finding that Ribas-Corbera and Haskell, alone or in combination, teach or
suggest the limitation:
6
Appeal2015-001244
Application 13/197 ,211
utilize a second QP value for at least one of U and V color
channels of the color video image for decoding said first
macroblock, wherein said second QP value is determined by at
least applying the bias value to the first QP value to provide a
first intermediate result,
as recited in claim 1, and as similarly recited in claims 6 and 12, and (2) in
concluding that claims 1, 6, and 12 are obvious under 35 U.S.C. § 103(a)
based on the combination ofNaimpally, Ribas-Corbera, and Haskell. Thus,
we do not sustain the Examiner's rejection of claims 1, 6, and 12, as well as
dependent claims 2, 3, 7-11, and 13-18, which variously depend therefrom.
DECISION
We reverse the Examiner's decision rejecting claims 1-3 and 6-18
under 35 U.S.C. § 103(a).
REVERSED
7