Ex Parte Jeon

Board of Patent Appeals and Interferences

Dec 28, 2011

10335331 (B.P.A.I. Dec. 28, 2011)

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.
10/335,331 12/31/2002 Byeong Moon Jeon 6111-000006/US 7207
30593 7590 12/29/2011
HARNESS, DICKEY & PIERCE, P.L.C.
P.O. BOX 8910
RESTON, VA 20195
EXAMINER
HALLENBECK-HUBER, JEREMIAH CHARLES
ART UNIT PAPER NUMBER
2486
MAIL DATE DELIVERY MODE
12/29/2011 PAPER
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.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________________

Ex parte BYEONG MOON JEON
____________________

Appeal 2009-012013
Application 10/335,331
Technology Center 2400
____________________

Before CARLA M. KRIVAK, CARL W. WHITEHEAD, JR., and
BRADLEY W. BAUMEISTER, Administrative Patent Judges.

KRIVAK, Administrative Patent Judge.

DECISION ON APPEAL
Appellant appeals under 35 U.S.C. § 134(a) from a final rejection of
claims 21, 26, and 28-38. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.

Appeal 2009-012013
Application 10/335,331

2
THE INVENTION
Appellant’s claimed invention relates to a moving picture coding
system (Spec. ¶ [0003]).
Independent claim 21, reproduced below, is representative of the
subject matter on appeal:
21. A method of predicting a bi-predictive block of a current
picture, the method comprising:

deriving a first temporal distance from a difference between a
picture order count (POC) for the current picture and a POC for
a first reference picture;

deriving a second temporal distance from a difference between
the POC for the first reference picture and a POC for a second
reference picture;

calculating first and second variable weight factors for first and
second motion-compensated blocks associated with the first and
second reference pictures, based on the first and second
temporal distances; and

predicting the bi-predictive block by applying the first and
second variable weight factors to the first and second motion-
compensated blocks, respectively.

REJECTIONS
The Examiner rejected claims 21, 26, and 28-36 under 35 U.S.C.
§ 102(e) as anticipated by Demos (US 6,816,552).
The Examiner rejected claims 37 and 38 under 35 U.S.C. § 103(a)
based upon the teachings of “Joint Video Team (JVT) of ISO/IEC MPEG
and ITU-T VCEG, Working Draft Number 2, January 29-February 1, 2002,
pp. 1-106” (“WD2”) and “ITU – Telecommunications Standardization
Appeal 2009-012013
Application 10/335,331

3
Sector, Video Coding Experts Group (Question 15), August 2000, pp. 1-2”
(“Q15”).

ANALYSIS
Rejection under § 102
The Examiner finds Demos discloses a method for predicting a bi-
predictive block of a current picture, including “deriving a first temporal
distance from a difference between a picture order count (POC) for the
current picture and a POC for a first reference picture,” and “deriving a
second temporal distance from a difference between the POC for the first
reference picture and a POC for a second reference picture,” as recited in
claim 21 (Ans. 3, 11).
Appellant contends Demos’s motion vector weightings cannot
reasonably be interpreted as picture order counts (App. Br. 18-19).
1

Appellant also contends Demos’s value “M” is fixed, and therefore is not a
temporal distance derived from picture order count numbers (App. Br. 19-
20). Further, Appellant argues, Demos’ system has no need to derive
temporal distances based on picture order counts because, for a given value
of M, it automatically knows the motion vector weightings (App. Br. 20,
22).
Appellant’s argument that Demos’s motion vector weightings are not
picture order counts misapprehends the Examiner’s position (App. Br. 18).
The Examiner does not assert the weightings are picture order counts.
Rather, the Examiner relies on the order of a picture relative to a particular

1
Appellant’s Appeal Brief filed April 28, 2008, is referenced throughout the
opinion.
Appeal 2009-012013
Application 10/335,331

4
reference picture for the limitation of a picture order count. (App. Br. 10-
11). The Examiner finds claim 21 does not require the picture order count
be explicit or be counted from any specific starting point, which is a
reasonable interpretation consistent with Appellant’s definition of “a count
value representing the order of a picture in a sequence of pictures” (Ans. 10-
11; App. Br. 17-18). That is, claim 21 recites deriving a temporal distance
from a difference between picture order counts regardless of the actual value
of the counts. Accordingly, the Examiner finds the difference between a
picture order count for the first B picture 201a in Demos’s Figure 2 and a
picture order count for the reference P or I picture A is one, because there is
a difference of one picture between them. For the second B picture 201b,
the difference is two. (Ans. 11).
Appellant’s argument that M is not a temporal distance derived from
picture order count numbers is not persuasive (App. Br. 19-20). Demos’s
value M is based on the count of pictures between two reference pictures.
As shown in Demos’s Figure 2, where M=3, the count of pictures from the
first reference picture A to the second reference picture B is three.
Additionally, M is a temporal distance because there is a temporal
relationship between pictures based on the frame rate (Ans. 11). For
example, Demos discloses at 24 frames per second (the rate of film movies)
there is a 1/24 second time distance between frames (pictures) (Demos, col.
4, ll. 5-6). Further, whether M is a fixed value does not impact the analysis
(App. Br. 20). As the Examiner asserts, claim 21 does not require uniquely
deriving an M value for each bi-predictive block (Ans. 12) and does not
recite “the temporal distances between the first and second reference picture
for different bi-predictive blocks may be different,” as alleged by Appellant
Appeal 2009-012013
Application 10/335,331

5
(App. Br. 20). Rather, claim 21 only recites predicting a single bi-predictive
block.
Appellant’s argument that Demos’ system automatically knows the
weightings to apply in interpolating a bi-predictive block is without merit
(App. Br. 20, 22). In fact, Appellant states “knowledge of whether the B
frame is the first following the P (or I) frame is required” in determining the
weightings (App. Br. 20). Thus, at least the position of the current B picture
relative to the previous P or I reference picture (which the Examiner finds is
equivalent to the claimed first temporal distance) is factored into the
weightings, as Appellant acknowledges (Ans. 12). The Examiner
additionally finds the value M (the second temporal distance) is also used to
determine the weightings (Ans. 4, 13).
Demos, therefore, teaches all the elements of claim 21 and is, thus,
anticipated. Claims 26 and 28-36, which depend from claim 21 and are not
separately argued, are also anticipated.
Rejection under § 103
With respect to claim 38, the Examiner finds WD2 discloses using a
difference in picture number (DPN) value to determine the number of a
picture in question (PNQ) relative to the number of a current picture (PNC),
wherein the DPN is a temporal distance (Ans. 7, 15; WD2 § 4.3.14.2). The
Examiner finds WD2 does not disclose applying variable weight factors to
predict a bi-predictive block (Ans. 7-8, 14). The Examiner relies on Q15 for
disclosing variable weight factors based on temporal distances for use in
predicting a bi-predictive block (Ans. 8). It would have been obvious, the
Examiner finds, to calculate variable weight factors for predicting bi-
Appeal 2009-012013
Application 10/335,331

6
predictive blocks in the encoding framework of WD2 to improve video
quality (Ans. 8).
Appellant contends WD2 does not disclose determining temporal
relationships based on picture numbers (App. Br. 26). Although WD2
discloses finding a DPN, Appellant argues this is different than deriving a
temporal distance from a difference in picture order counts (App. Br. 27-29).
Appellant also contends Q15 does not disclose determining weight factors
for predicting a bi-predictive block based on differences in picture order
counts (App. Br. 29). Additionally, Appellant argues, even if WD2’s DPN
is a temporal distance, the Examiner has not provided a rationale for using
DPNs in deriving the temporal distances Q15 requires for determining
weight factors, and thus, the combination is based on improper hindsight
(App. Br. 33).
As the Examiner finds, WD2’s DPN is a temporal distance because it
represents a difference in picture numbers (PNs), which WD2 defines as a
value incrementally increased by one for each picture relative to the
temporally previous picture in bit-stream order (Ans. 14-15; WD2 § 4.3.11).
The Examiner also explains that, in combining Q15’s improvement with the
existing coding standard of WD2, one would use WD2’s DPN values for
Q15’s required temporal distances because the DPN values are already
available in WD2’s encoding framework (Ans. 14-15). Although Appellant
argues there is no rationale for combining Q15 with WD2, Appellant has not
addressed the Examiner’s stated reason for combining the references,
namely, improving video quality through the use of a temporally weighted
prediction of bi-predictive blocks (Ans. 8).
Appeal 2009-012013
Application 10/335,331

7
Therefore, claims 38 and 37, which fall together, are obvious over
WD2 and Q15 (See App. Br. 15).

DECISION
The Examiner’s decision rejecting claims 21, 26, and 28-38 is
affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED

kis