Ex Parte Chevance et al

Board of Patent Appeals and Interferences

Apr 9, 2008

09786432 (B.P.A.I. Apr. 9, 2008)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

Ex parte CHRISTOPHE CHEVANCE,
PIERRE RUELLOU, and DOMINIQUE THOREAU
____________

Appeal 2007-3604
Application 09/786,432
Technology Center 2600
____________

Decided: April 9, 2008
____________

Before ANITA PELLMAN GROSS, JOHN A. JEFFERY, and KARL
EASTHOM, Administrative Patent Judges.

JEFFERY, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134 from the Examiner’s
rejection of claims 1-4, 7, 10, and 11. Claims 5, 6, 8, 9, 12 and 13 have been
indicated as containing allowable subject matter (Ans. 10-11; App. Br. 2).
We have jurisdiction under 35 U.S.C. § 6(b). We reverse.
Appeal 2007-3604
Application 09/786,432

STATEMENT OF THE CASE
Appellants invented a method of motion estimation for a series of
images. The process includes the step of segmentation of a video image into
blocks. A histogram is generated representing the horizontal and vertical
components of the movement vectors associated with the image of the
series. Predominant vectors within the histogram are located using various
techniques. One of the predominant vectors is then reassigned to the blocks.
The reassignment can be enhanced by scaling the vectors using spatial and
temporal filtering.1 Claim 1 is illustrative:
1. Method of movement estimation for a sequence of images
including segmentation of a current video image into image blocks,
movement estimation per image block between the current image
and a previous image in order to obtain a movement vector field
for said current image, a stage of reassignment of a vector to a
block by selecting one movement vector from among N
predominant vectors, wherein the predominant vectors are the ones
of the group of vectors belonging to the movement vector field for
said current image and at least to the movement vector field for a
preceding image corresponding to a movement vector field
between said preceding image and a further preceding image, the
vectors being scaled according to the temporal distance to which
they correspond.

The Examiner relies on the following prior art references to show
unpatentability:
Kerdranvrat US 5,193,001 Mar. 9, 1993

Lee US 6,317,460 B1 Nov. 13, 2001
(filed May 12, 1998)

1 See generally Spec. 1:25-35 and 3:11-5:7.
2
Appeal 2007-3604
Application 09/786,432

The Examiner’s rejection is as follows:
1. Claims 1-4, 7, 10, and 11 stand rejected under 35 U.S.C. § 103(a)
as unpatentable over Kerdranvrat and Lee.
Rather than repeat the arguments of Appellants or the Examiner, we
refer to the Brief2 and the Answer3 for their respective details. In this
decision, we have considered only those arguments actually made by
Appellants. Arguments, which Appellants could have made but did not
make in the Brief, have not been considered and are deemed to be waived.
See 37 C.F.R. § 41.37(c)(1)(vii).

OPINION
In rejecting claims under 35 U.S.C. § 103, it is incumbent upon the
Examiner to establish a factual basis to support the legal conclusion of
obviousness. See In re Fine, 837 F.2d 1071, 1073 (Fed. Cir. 1988). In so
doing, the Examiner must make the factual determinations set forth in
Graham v. John Deere Co., 383 U.S. 1, 17 (1966).
Discussing the question of obviousness of a patent that claims a
combination of known elements, the Court in KSR Int’l v. Teleflex, Inc., 127
S. Ct. 1727 (2007) explains:
When a work is available in one field of endeavor, design
incentives and other market forces can prompt variations of it,
either in the same field or a different one. If a person of
ordinary skill can implement a predictable variation, § 103

2 We refer to the most recent Appeal Brief filed August 22, 2006 throughout
this opinion.
3 We refer to the Examiner’s Answer mailed November 17, 2006 throughout
this opinion.
3
Appeal 2007-3604
Application 09/786,432

likely bars its patentability. For the same reason, if a technique
has been used to improve one device, and a person of ordinary
skill in the art would recognize that it would improve similar
devices in the same way, using the technique is obvious unless
its actual application is beyond his or her skill. Sakraida [v. AG
Pro, Inc., 425 U.S. 273 (1976)] and Anderson's-Black Rock[,
Inc. v. Pavement Salvage Co., 396 U.S. 57 (1969)] are
illustrative—a court must ask whether the improvement is more
than the predictable use of prior art elements according to their
established functions.
KSR, 127 S. Ct. at 1740. If the claimed subject matter cannot be fairly
characterized as involving the simple substitution of one known element for
another or the mere application of a known technique to a piece of prior art
ready for the improvement, a holding of obviousness can be based on a
showing that “there was an apparent reason to combine the known elements
in the fashion claimed.” Id. at 1740-41. Such a showing requires “some
articulated reasoning with some rational underpinning to support the legal
conclusion of obviousness. . . . [H]owever, the analysis need not seek out
precise teachings directed to the specific subject matter of the challenged
claim, for a court can take account of the inferences and creative steps that a
person of ordinary skill in the art would employ.” Id. at 1741.
If the Examiner’s burden is met, the burden then shifts to the
Appellants to overcome the prima facie case with argument and/or evidence.
Obviousness is then determined on the basis of the evidence as a whole and
the relative persuasiveness of the arguments. See In re Oetiker, 977 F.2d
1443, 1445 (Fed. Cir. 1992).
Regarding independent claim 1, the Examiner's rejection essentially
finds that Kerdranvrat discloses a method of movement estimation for a
sequence of images with all the features recited in claim 1, except for the
4
Appeal 2007-3604
Application 09/786,432

step of scaling the vectors according to the temporal distance as claimed.
The Examiner cites Lee to teach this missing limitation. In view of Lee’s
teaching, the Examiner concludes that it would have been obvious to one of
ordinary skill in the art at the time of the invention to apply a temporal
scaling technique to the predominant vectors of Kerdranvrat “in order to
reduce the processing load” (Ans. 4).
Appellants mainly argue that Kerdranvrat fails to disclose or teach the
limitation:
a stage of reassignment of a vector to a block by selecting one
movement vector from among N predominant vectors, wherein the
predominant vectors are the ones of the group of vectors belonging to
the movement vector field for said current image and at least to the
movement vector field for a preceding image

(App. Br. 8-9). Appellants state that the claimed invention selects a
predominant vector from two movement fields, while Kerdranvrat selects a
predominant vector from one movement field (App. Br. 8-9).
The issue before us is whether Kerdranvrat discloses, teaches, or
suggests the N predominant vectors used to select the reassigned movement
vector for a block are from the group of vectors belonging to: (1) the
movement vector field of the current image and (2) the movement vector
field for the preceding image. For the following reasons, we find that
Kerdranvrat does not.
Kerdranvrat discloses a method of movement estimation for a
sequence of images. The method includes the steps of segmentation of a
video image into image blocks and reassignment of a vector to each block by
selecting one movement vector from among N predominant vectors. The Nt
dominant vectors in Kerdranvrat are located through a series of steps,
5
Appeal 2007-3604
Application 09/786,432

including extraction from a table containing vector values from the current
image and elimination of secondary maxima. Kerdranvrat, then, includes a
filtering step and arrives at N predominant vectors (Kerdranvrat, col. 2, ll.
36-39 and 44-47, col. 3, ll. 31-39, col. 5, l. 12 - col. 7, l. 41; Figs. 1 and 5-
8b).
The later filtering step of Kerdranvrat, in our view, does not result in
the N predominant vectors being from a group of vectors belonging to: (1)
the movement vector field for said current image and (2) the movement
vector field for the preceding image. At best, the step involves a comparison
of the dominant vectors from the current image with the dominant vectors
extracted from the previous image. MEM1 contains the Nmax vectors or
vectors per one image quarter from the current image, and MEM2 contains
the dominant vectors from the previous image. When one of the vectors in
MEM1 is distant from all the vectors in MEM2, this vector is eliminated
from the table SORT. There is no discussion of transferring the vectors in
MEM2 into MEM1 or the table SORT. In the end, while a comparison is
made between vectors of the current and previous images, only the Nmax
vectors close to the vectors in MEM2 or N predominant vectors from the
current image are sent to the next module for reassignment (Kerdranvrat,
col. 7, ll. 11-41; Figs. 8a-b and 11).
The Examiner argues “after the SORT table is time filtered, both its
predominant vectors (current image), and the dominant vectors from the
MEM2 table (preceding image) are transferred to the reassignment module”
(Ans. 7). We do not find support for this interpretation. Kerdranvrat
discloses the Nmax vectors from the current image, not the vectors from the
MEM2 table, are sent to and used in the reassignment module (Kerdranvrat,
6
Appeal 2007-3604
Application 09/786,432

col. 7, ll. 33-36 and 44-50). When looking at this passage of Kerdranvrat in
its entirety, “these vectors” discussed in column 7, line 35 of Kerdranvrat as
being transmitted to the next module refer to the vectors in the table SORT.
To support this interpretation, Kerdranvrat further discloses the values from
the table SORT, not MEM2, are used in calculating the reassigned values for
each block (Kerdranvrat, col. 7, l. 55- col. 8, l. 10). This indicates the blocks
are reassigned vectors from only the table SORT or vectors from the current
image.
The Examiner admits that the vector values in the table SORT are
close but different from the vector values in MEM2 (Ans. 7). In our view, if
a block was reassigned a movement vector from among vectors of MEM2 or
a previous image, then the block would have a reassigned value that differs
from the predominant vectors in the table SORT, as described by
Kerdranvrat, and would be contrary to Kerdranvrat’s teachings (Kerdranvrat,
col. 7, l. 55 – col. 8, l. 10; Fig. 11). As stated previously, Kerdranvrat uses
only Nmax vector values from the current image that are in the table SORT
when reassigning the blocks.
Based on the above findings, we conclude that Kerdranvrat does not
disclose or reasonably teach the predominant vectors used in the
reassignment step are from the group of vectors belonging to the movement
vector field of the current image and the movement vector field of the
preceding image as recited in claim 1.
Appellants also argue that Lee does not provide a teaching for this
missing limitation (App. Br. 9-10). Lee discloses a method for movement
estimation of a sequence of images in order to reduce computational
requirements (Lee, col. 1, ll. 14-16 and col. 3, ll. 53-58) and discusses
7
Appeal 2007-3604
Application 09/786,432

assigning blocks motion vectors based on a block-based approach and other
conventional techniques, such as SAD, SSD, or MAD (Lee, col. 1, l. 60 -
col. 2, l. 13 and col. 6, ll. 7-11). However, Lee does not describe any details
regarding the reassignment of the motion vectors, and in particular, there is
no discussion of a reassignment from among N predominant vectors chosen
from the group of vectors belonging to the movement vector field of the
current image and the movement vector field of the preceding image. As
such, Lee does not disclose, teach, or suggest the above-discussed missing
limitation.
As neither reference discloses or teaches the step of reassigning a
vector to a block from among N predominant vectors chosen from the group
of vectors belonging to the movement vector field of the current image and
the movement vector field of the preceding image, Kerdranvrat, Lee, and
their combination do not provide a persuasive teaching of reassigning a
vector to a block by selecting one movement vector from among N
predominant vectors, wherein the predominant vectors are the ones of the
group of vectors belonging to the movement vector field for the current
image and at least to the movement vector field for a preceding image as
recited in claim 1.
For the above reasons, we will not sustain the Examiner’s rejection of
independent claim 1 nor dependent claims 2-4, 7, 10, and 11 for similar
reasons.
DECISION
We have not sustained the Examiner's rejections with respect to any of
the claims on appeal. Therefore, the Examiner’s decision rejecting claims 1-
4, 7, 10, and 11 is reversed.
8
Appeal 2007-3604
Application 09/786,432

REVERSED

eld

Joseph S Tripolis
Thomson Multimedia Licensing Inc
P.O. Box 5312
Princeton, NJ 08543-5312
9