Uniloc 2017 LLC

Patent Trials and Appeals Board

Apr 20, 2021

IPR2020-00044 (P.T.A.B. Apr. 20, 2021)

Trials@uspto.gov Paper 24
571-272-7822 Date: April 20, 2021

UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE PATENT TRIAL AND APPEAL BOARD

NETFLIX, INC.,
Petitioner,
v.
UNILOC 2017 LLC,
Patent Owner.

IPR2020-00044
Patent 6,584,229 B1

Before SALLY C. MEDLEY, BRYAN F. MOORE, and
IFTIKHAR AHMED, Administrative Patent Judges.
MOORE, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining All Challenged Claims Unpatentable
35 U.S.C. § 318(a)

I. INTRODUCTION
A. Background
Netflix, Inc. (“Petitioner” or “Netflix”) filed a Petition requesting
inter partes review of claims 1, 3, 5, and 6 (the “challenged claims”) of U.S.
IPR2020-00044
Patent 6,584,229 B1
2
Patent No. 6,584,229 B1 (Ex. 1001, the “’229 patent”). Paper 1 (“Pet.”) at
6. Uniloc 2017 LLC (“Patent Owner”) filed a Preliminary Response. Paper
6 (“Prelim. Resp.”). On April 27, 2020, we instituted trial. Paper 7 (“Inst.
Dec.”). Patent Owner filed a Response. Paper 231 (“PO Resp.”). Petitioner
filed a Reply. Paper 12 (“Reply”). Patent Owner filed a Sur-Reply. Paper
16 (“Sur-Reply”). An oral argument was held on February 3, 2021, and a
transcript was entered into the record. Paper 22 (“Tr.”).
We have jurisdiction to conduct this inter partes review under
35 U.S.C. § 6. This Final Written Decision is issued pursuant to 35 U.S.C.
§ 318(a) and 37 C.F.R. § 42.73. For the reasons discussed herein, we
determine that Petitioner has shown, by a preponderance of the evidence,
that claims 1, 3, 5 and 6 of the ’229 patent are unpatentable.
B. Related Proceedings
The parties identify the following pending district court proceeding
involving the ’229 patent: Uniloc 2017 LLC v. Netflix, Inc., No. 8:18-cv-
02055 (CD Cal.). Pet. 4; Paper 4, 2.
C. Real Parties-in-Interest
Petitioner identifies Netflix, Inc. as the real party-in-interest. Pet. 3.
Patent Owner identifies Uniloc 2017 LLC as the real party-in-interest. Paper
4, 1.
D. The ’229 Patent
The ’229 patent describes a macroblock-based object-oriented coding
method of a video (e.g., moving) image sequence that can be coded at a
relatively low coding bit rate. Ex. 1001, 1:66–2:2. The image sequence is

1 The Patent Owner Response was filed as Paper 10 without page numbers.
Patent Owner refiled an authorized Patent Owner Response with page
numbers as Paper 23.
IPR2020-00044
Patent 6,584,229 B1
3
split into a background region and a foreground object region and encoded
into macroblock-based units in a manner that achieves the relatively low
coding bit rate. Id. at 2:2–8. Figure 1, reproduced below, is a flow chart of
the invention.

IPR2020-00044
Patent 6,584,229 B1
4
Figure 1, above, is a high level flow chart of the invention. Id. at 2:29–32.
E. Illustrative Claim
The ’229 patent has six claims. Four claims (1, 3, 5 and 6) are
challenged in the Petition. See supra. Of the challenged claims, only claim
1 is independent. Claim 1 recites:
1. A method, for use in an macroblock-based object
oriented coding of a image signal, wherein the image signal has
a stationary background region and an object region and contains
a current frame and a previous frame, each frame including a
plurality of macroblocks, comprising the steps of:
a) dividing the stationary background region and the
object region from an inputted video in a macroblock-by-
macroblock basis by using a difference between the previous
frame and the current frame;
b) coding shape information of the object region by using
a known coding technique to generate coded shape information;
c) coding pixel information of each macroblock contained
in the object region by using a selected known coding technique
to generate coded object pixel information;
d) generating coded pixel information of a previous frame
macroblock corresponding to each current frame macroblock
contained in the stationary background region as coded
stationary pixel information; and
e) storing or transmitting coded data coded shape
information, coded object pixel information and coded stationary
pixel information as coded image signal; and
wherein the step d) includes the step of reusing
corresponding coded pixel information macroblock contained in
the previous frame without coding the pixel information of each
macroblock contained in the current frame when a difference
between a pixel value of the macroblock of the current frame and
that of the macroblock of the previous frame in the same position
is identical to or smaller than a predetermined threshold value.
Ex. 1001, 3:40–4:21.
IPR2020-00044
Patent 6,584,229 B1
5
F. Evidence
The Petition relies on the following references:
Name Reference Exhibit
Talluri Talluri, R., A robust, scalable, object-based
video compression technique for very low bit-
rate coding, IEEE Transactions on Circuits and
Systems for Video Technology, 7(1), 221–233,
published Feb. 1997
1003
Lin US 6,192,148 B1, issued Feb 20, 2001 (filed
Nov. 5, 1998)
1006
Monro WO 98/11730 A1, filed Sept 12, 1997,
published Mar. 19, 1998
1007
Chai Chai, D., Foreground/background video coding
scheme, Proceedings of 1997 IEEE
International Symposium on Circuits and
Systems. Circuits and Systems in the
Information Age ISCAS 97, published June
1997
1008
Pet. 5–6.
In addition, Petitioner submits the Declaration of Eli Saber, Ph.D.
(Ex. 1002). Patent Owner has not submitted an expert declaration.
G. Asserted Grounds of Unpatentability
Petitioner asserts the challenged claims are unpatentable on the
following grounds.
Claims Challenged 35 U.S.C. §2 References/Basis
1, 3, 5, 6 103 Talluri, Lin
1, 5, 6 103 Monro, Chai
Pet. 6.

2 The Leahy-Smith America Invents Act, Pub. L. No. 112-29, 125 Stat. 284
(2011) (“AIA”), amended several provisions of 35 U.S.C., including § 103.
Because the ’229 patent has an effective filing date before the effective date
of the applicable AIA amendments, we refer to the pre-AIA version of
35 U.S.C. § 103.
IPR2020-00044
Patent 6,584,229 B1
6
II. PRELIMINARY MATTERS
A. Level of Ordinary Skill
Petitioner contends:
As of the Aug. 30, 1999 priority date of the ’229 patent, a [person
of ordinary skill in the art (“POSA”)] would have had a
bachelor’s degree in electrical engineering, computer
engineering, computer science, or a similar field with at least two
years of experience with digital video, coding and/or processing.
. . . A person with less education but more relevant practical
experience may also meet this standard. A POSA would have
had basic familiarity with the major MPEG and ITU-T video
coding standards and encoding techniques used in the standards.

Pet. 27–28 (citing Ex. 1002 ¶ 21).
“Patent Owner does not offer a competing definition for” a person of
ordinary skill in the art. PO Resp. 6. We regard Petitioner’s definition as
consistent with the prior art before us. See Okajima v. Bourdeau, 261 F.3d
1350, 1355 (Fed. Cir. 2001) (prior art itself may reflect an appropriate level
of skill). Thus, we adopt Petitioner’s proposal.
B. Claim Construction
Because the Petition was filed after November 13, 2018, we construe
the challenged claims by applying the standard used in federal courts, in
other words, “the claim construction standard that would be used to construe
the claim in a civil action under 35 U.S.C. [§] 282(b),” which is articulated
in Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc). 37
C.F.R. § 42.100(b) (2019). Further, “[a]ny prior claim construction
determination concerning a term of the claim in a civil action, or
IPR2020-00044
Patent 6,584,229 B1
7
a proceeding before the International Trade Commission, that is timely made
of record in the inter partes review proceeding will be considered.” Id.
Under the standard set forth in Phillips, 415 F.3d at 1312–19, claim
terms are given their ordinary and customary meaning, as would have been
understood by a person of ordinary skill in the art at the time of the
invention, in light of the language of the claims, the specification, and the
prosecution history of record. See Thorner v. Sony Comput. Entm’t Am.
LLC, 669 F.3d 1362, 1365–66 (Fed. Cir. 2012). There is a “heavy
presumption,” however, that a claim term carries its ordinary and customary
meaning. CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1366 (Fed.
Cir. 2002) (citation omitted).
We are also guided by the principle that we only construe claim terms
if, and to the extent that, it is necessary for the purpose of the proceeding.
See, e.g., Wellman, Inc. v. Eastman Chem. Co., 642 F.3d 1355, 1361 (Fed.
Cir. 2011) (“[C]laim terms need only be construed ‘to the extent necessary
to resolve the controversy.’”) (quoting Vivid Techs., Inc. v. Am. Sci. &
Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)).
We address four limitations below.
“dividing the stationary background region and the object region from an
inputted video in a macroblock-by-macroblock basis by using a difference
between the previous frame and the current frame”

Patent Owner requests that we construe step a) of claim 1, i.e.
dividing the stationary background region and the object region from an
inputted video in a macroblock-by-macroblock basis by using a difference
between the previous frame and the current frame. PO Resp. 11.
Specifically, during its non-obviousness discussion, Patent Owner argues
IPR2020-00044
Patent 6,584,229 B1
8
that the claim language “in a macroblock-by-macroblock basis” limits how
the division must be done. Id.
Petitioner responds, “[n]othing in the language of the ‘dividing’ step
mandates a single step approach,” and “[a]s long as the dividing is done on a
macroblock-by-macroblock basis and ‘a difference between the previous
frame and the current frame’ is used, the limitation is satisfied. Reply 8.
Additionally Petitioner’s declarant testifies that “[t]he claim language of step
‘a)’ describes what is done: ‘dividing the stationary background region and
the object region from an inputted video in a macroblock-by-macroblock
basis,’ and provides one requirement as to how, i.e., ‘using a difference
between the previous frame and the current frame.’” Ex. 1027 ¶ 213; Reply
8. We credit Petitioner’s declarant’s testimony on claim construction. Ex.
1027 ¶ 213. The technology in this IPR is complex and the specification is
sparse. In complex cases, where obviousness is asserted as a ground of
unpatentability, “expert testimony may be critical.” Wyers v. Master Lock
Co., 616 F.3d 1231, 1240 n.5 (Fed. Cir. 2010) (citations omitted).
To resolve these two readings of the claim language, we look to the
Specification. The Specification contains one paragraph describing the
dividing process. That paragraph starts by reciting that “the video signal is
divided into a macroblock-based stationary back ground region and a
macroblock-based moving object region by using the difference between a
previous frame of the input video signal and a current frame of the input
video at step 10.” Ex. 1001, 2:45–49.3 In other words, the Specification
describes dividing the video signal into macroblock-based “regions” “by

3 The rest of the paragraph describes a specific method of dividing which is
covered by claim 2. Ex. 1001, 2:50–62. Claim 2 is not challenged in this
IPR. Pet. 6.
IPR2020-00044
Patent 6,584,229 B1
9
using” the difference between a previous frame of the input video signal.
See Ex. 1001, 45–49. A fair reading of this sentence, consistent with
Petitioner’s contention, suggests that the how of this process is by using the
“difference” and the what (i.e., the result) of the process is creation of
macroblock based regions. Thus, the “dividing the stationary background
region and the object region from an inputted video in a macroblock-by-
macroblock basis” can be read to mean the dividing is done into a
macroblock “based” stationary background region and a macroblock
“based” object region using the “difference.” Thus, we adopt a construction
that “dividing the stationary background region and the object region from
an inputted video in a macroblock-by-macroblock basis by using a
difference between the previous frame and the current frame” means “using
a difference between the previous frame and the current frame such that the
stationary background region and the object region from an inputted video
are divided on a macroblock-by-macroblock basis.”
Patent Owner also argues that the preamble of claim 1 provides
antecedent basis to this claim limitation. PO Resp. 11. Specifically, the
preamble of claim 1 recites “the image signal . . . contains a current frame
and a previous frame, each frame including a plurality of macroblocks.” Ex.
1001, 3:42–46 (emphasis added). Here Patent Owner argues:
A plain reading of the claim language also reveals that the
“dividing” step relies upon the macroblocks of a given frame to
be already defined. This understanding is at least reflected in the
definitional recitation that the “dividing” step must itself be
performed “in a macroblock-by-macroblock basis.” This
understanding is also reflected in the definitional statements
recited in the preamble that the “current frame” and “previous
frame” must each include a plurality of macroblocks. The
“dividing” step of claim 1 derives antecedent basis from this
definitional statement in the preamble by referring to “the
IPR2020-00044
Patent 6,584,229 B1
10
previous frame and the current frame” and respective
macroblocks thereof.

PO Resp. 11. Thus, Patent Owner argues that the fact that the input “current
frame” and “previous frame” each include a plurality of macroblocks means
that the input to the “dividing” is already in macroblocks so the dividing
must be done on a macroblock-by-macroblock basis. Id.
It is true that the limitation at issue in the body of claim 1 recites “by
using a difference between the previous frame and the current frame.” And
although we agree with Patent Owner that the previous frame and the
current frame in this limitation refer back to the preamble, Patent Owner
fails to explain the significance of the preamble to the dividing step. For
example, Patent Owner does not explain sufficiently how the preamble
provides antecedent basis for “in a macroblock-by-macroblock basis,” and
we find none. Instead, we read that limitation to mean the dividing is done
into a macroblock “based” stationary background region and a macroblock
“based” object region using the “difference.”
In fact, although Patent Owner suggests the claim construction in the
related District Court proceeding supports its position in this IPR (Sur-Reply
7), the District Court’s decision resolved the issue whether the term “in a
macroblock-by-macroblock basis” modifies the “inputted video” or the
“dividing” by siding with Netflix (i.e., Petitioner) and deciding that the term
“in a macroblock-by-macroblock basis” modifies dividing. Ex. 1030, 25,
26.
Nevertheless, the main issue in this IPR, i.e. whether the initial
dividing—in a two-step process—must be on a macroblock-by-macroblock
basis or the result of the dividing must be on a macroblock-by-macroblock
basis was not at issue in the District Court. The District Court construed this
IPR2020-00044
Patent 6,584,229 B1
11
limitation as “the stationary background region and the object region from
an inputted video are divided on a macroblock-by-macroblock basis,” but in
doing so it recognized in doing so that it was supporting Petitioner’s
construction which was “dividing into regions that are defined as one or
more macroblocks.” Id. at 26. We find the District Court’s use of the
phrase “are divided” suggests that the District Court was referring to the
result of the dividing as proposed by Petitioner. Additionally, the District
Court makes clear that it is distinguishing between dividing on a pixel basis
versus a macroblock basis. Id. (“dividing a video not by pixel unit but by
macroblock-unit.”) (quoting Patent Owner’s argument before the District
Court); see also Tr. 14:25–15:12. Thus, we determine that our construction
is consistent with the construction in the District Court, i.e. using a
difference between the previous frame and the current frame such that the
stationary background region and the object region from an inputted video
are divided on a macroblock-by-macroblock basis.
“generating coded pixel information of a previous frame macroblock”
Petitioner requests that we construe step d) of claim 1 (“generating
coded pixel information of a previous frame macroblock . . .”) in one of two
alleged possible alternatives, i.e., step d) refers to “(1) encoding
macroblocks of the previous frame, or (2) encoding macroblocks of the
current frame.” Pet. 24 (citing Ex. 1002 ¶¶ 80–85). Nevertheless, Petitioner
asserts, “[u]ltimately which interpretation is correct is of little consequence
because the prior art clearly discloses encoding both the previous frame and
the current frame, making the step obvious under either interpretation.” Id.
Patent Owner does not respond but asserts “regardless whether the Board
adopts the constructions expressly proposed in the Petition, . . . the claim
IPR2020-00044
Patent 6,584,229 B1
12
construction proposals set forth in the Petition do not present dispositive
issues here.” PO Resp. 7.
Patent Owner has not offered a construction. See generally PO Resp.
Because we find below that Petitioner has shown step d) of claim 1
sufficiently under either construction that Petitioner proposes, we decline to
construe step d) of claim 1. Pet. 41–44, 66–68; see PO Resp. 19–24
(addressing the second clause of step e) (“reusing limitation”) of claim 1,
but not addressing whether step d) refers to encoding macroblocks of the
previous frame or the current frame).
“coded data coded shape information”
Petitioner also requests that we construe “coded data coded shape
information, coded object pixel information and coded stationary pixel
information” as if it read “coded data including coded shape information,
coded object pixel information and coded stationary pixel information.”
Pet. 27. Patent Owner does not respond. PO Resp. 7–8.
The United States Court of Appeals for the Federal Circuit has held
that a district court can act to correct an error in a patent by interpretation of
the patent where no certificate of correction has been issued if “(1) the
correction is not subject to reasonable debate based on consideration of the
claim language and the specification and (2) the prosecution history does not
suggest a different interpretation of the claims.” Novo Indus., L.P. v.
MicroMolds Corp. 350 F.3d 1348, 1354 (Fed. Circ. 2003) (“Novo Indus.”);
cf. Chef Am., Inc. v. Lamb-Weston, Inc., 358 F.3d 1371, 1374 (Fed. Cir.
2004) (“courts may not redraft claims, whether to make them operable or to
sustain their validity”). We have applied this same standard in our
proceedings and use our discretion to apply this standard here. E.g.,
Research Prods. Corp. v. Honeywell Int’l Inc., IPR2018-00976, Paper 6 at
IPR2020-00044
Patent 6,584,229 B1
13
13 (PTAB Nov. 7, 2018) (Institution Decision) (applying Novo Indus. and
holding “one of more programmable characters” was obvious error;
corrected to “one or more programmable characters”); see also Fitbit, Inc. v.
Valencell, Inc., 964 F.3d 1112, 1119–20 (Fed. Cir. 2020) (suggesting the
standard from Novo Indus. extends to corrections the Board may make in
post-grant proceedings, and holding it was error to not correct a
“conspicuous” and undisputed error related to antecedent basis); see e.g.,
Apple Inc. v. Achates Reference Publ’g, Inc., IPR2013-00080, Paper 22 at
10–12 (PTAB June 3, 2013).
According to Petitioner, because the claim phrase “coded data”
describes the collection of subsequent elements of the list, the omission of
the word “including” is a typographical error that one of ordinary skill
would recognize. Pet. 27 (citing Ex. 1002 ¶¶ 86–87). In support of this
argument, Petitioner asserts its construction is consistent with the
Specification’s “use of ‘coded data’ to refer generically to the encoder
output which is stored or transmitted. Id. (quoting Ex. 1001, Abstract
(‘storing or transmitting the coded data’), id. at 3:28–32 (‘Such coded data
through these processes can be stored at the media, e.g., CD-Rom, or
transmitted via a telecommunication network in accordance with the coding
machine’s purpose.’)).
We agree with Petitioner that there is a typographical error in the
claim. We find the claim appears to be missing a word such as “including”
after “coded data.” We determine that the omission of the word “including”
is the kind of error that is considered a drafting error and is not subject to
reasonable debate. We do not discern that the claim language was changed
during prosecution, nor was the apparent typographical error discussed. See
Ex. 1023, 1 (original claim); see also generally Ex. 1023 (no discussion of
IPR2020-00044
Patent 6,584,229 B1
14
apparent typographical error). Thus, we find treating the omission of the
word “including” as an error is consistent with the Specification.
Accordingly, we read the limitation “coded data coded shape information,
coded object pixel information and coded stationary pixel information” as if
it read “coded data including coded shape information, coded object pixel
information and coded stationary pixel information,” as Petitioner suggests.
“using a selected known coding technique”
Step c) of claim 1 recites “c) coding pixel information of each
macroblock contained in the object region by using a selected known coding
technique to generate coded object pixel information.” Patent Owner
construes this phrase, as part of a non-obviousness argument, by asserting
“the coding technique of step c) must have been ‘selected’ from among
available system options.” PO Resp. 9–10. Patent Owner argues, “[t]he
Petition is deficient on its face as failing to give any meaningful effect to the
‘selected’ qualifier in step c) of claim 1.” Id. at 8.
Step c) recites “coding pixel information of each macroblock
contained in the object region by using a selected known coding technique to
generate coded object pixel information.” In contrast, step b) of claim 1
recites “coding shape information of the object region by using a known
coding technique to generate coded shape information.” Thus, step b) does
not recite “selected,” which suggests, according to Patent Owner, that
“selected” in step c) must be given the meaning Patent Owner proposes. Id.
at 9–10. We disagree. Although the general rule is that the use of two
different terms in a patent claim implies that they have different
meanings, “that implication is overcome where . . . the evidence indicates
that the patentee used the two terms interchangeably.” Baran v. Med.
Device Techs., Inc., 616 F.3d 1309, 1316 (Fed. Cir. 2010). Additionally,
IPR2020-00044
Patent 6,584,229 B1
15
“[t]he preference for giving meaning to all terms . . . is not an inflexible rule
that supersedes all other principles of claim construction.” SimpleAir, Inc. v.
Sony Ericsson Mobile Commc’ns AB, 820 F.3d 419, 429 (Fed. Cir. 2016)
(citations omitted).
Patent Owner also relies on the Specification which recites that
“[a]fter the region distinction is completed, a coding technique is determined
for the macroblock belonging to each region at step 20.” Id. at 10. Patent
Owner asserts the fact “[t]hat a coding technique is determined ‘after the
region distinction is completed’ (id.) confirms that this determination or
selection is not based exclusively on the region distinction itself (e.g.,
whether a given region pertains to a stationary background or, instead, to a
moving object).” We disagree. The temporal reference to “after” does not
inform the basis upon which the selection is made. Additionally, because
the term “selected” is not the same as “determined,” it is not clear that the
reference to “determining” in the Specification refers to selecting from
available options as suggested by Patent Owner.
As Petitioner contends, there is no disclosure in the Specification of
selecting a coding technique from “available options.” Reply 3.
Reproduced below is Figure 1 of the ’229 patent.
IPR2020-00044
Patent 6,584,229 B1
16

Figure 1 of the ’229 patent, reproduced above, shows the steps of the
process. Ex. 1001, 2:29–32. Petitioner also contends that the only decision
point in the flow chart of the process is step 40—where it is determined
whether or not to encode a macro block at all. Id. at 4.
Although, step 40 does appear to be the only decision branch in the
chart, step 20 appears on its face to be relevant to the “selecting” in the
claim because it states “determin[ing]” “coding techniques.” Nevertheless,
the Specification’s description of step 20 is confusing. As described above,
the Specification states “[a]fter the region distinction is completed, a coding
technique is determined for the macroblock belonging to each region at step
IPR2020-00044
Patent 6,584,229 B1
17
20.” Ex. 1001, 2:63–67. To the contrary, however, in the same paragraph,
still ostensibly describing step 20, the Specification recites,
[w]hen this difference value is not bigger than the predetermined
threshold value, it is determined that the macroblock data of the
reconstructed previous frame is to be utilized. On the other hand,
when this difference value is bigger than the predetermined
threshold value, it is determined that the macroblock data of the
current frame is to be coded by well-known macroblock coding
technique.

Ex. 1001, 3:2–9. This description clearly refers to the determination being
made in step 40, and states when a macroblock needs to be encoded, “[a]
well-known macroblock coding technique” is used, therefore suggesting that
the coding technique is the one that was previously selected.
Curiously, a few lines down another paragraph begins by stating,
“[t]he process for determining whether the coding process of the pixel value
of the macroblock is to be executed or not is to be performed at step 40.”
(Ex. 1001, 3:20–22). The rest of that paragraph, however, does not describe
the process of determining whether or not to execute a coding process.
Thus, the Specification is lacks sufficient clarity and thus unhelpful in
resolving this issue.
Petitioner also contends that “[t]he term ‘selected’ in step c) is merely
an adjective qualifying the ‘known coding technique’ used to encode the
object region macroblocks.” Reply 3. We agree. Looking at the words of
the claim itself, we note that “selected” is not a verb, such as select or
selecting, that would normally denote a step in a method. Thus, we
determine that the limitation of “selected” is not a step in the asserted
method but, instead, is a characteristic of the known coding method, i.e., it is
the one that was previously selected. See Summit 6, LLC v. Samsung Elecs.
Co., 802 F.3d 1283, 1290–91 (Fed Cir. 2015) (finding claim limitation
IPR2020-00044
Patent 6,584,229 B1
18
was “not a step in the claimed method,” but was instead only “a phrase that
characterizes the claimed pre-processing parameters,” as the limitation
was “not used as a verb . . ., but instead is a part of a phrase that conveys
information about” claimed method parameters).
Additionally, nothing in the Specification, the claims, or any extrinsic
evidence suggest a plain meaning of “selected” such that a selection must be
“from available system options.” Id. at 1292. Given that the Specification
does not support sufficiently a finding that a selection “from available
options” is made and the fact that selected is an adjective rather than a verb,
we determine that “selected” simply qualifies the “known coding technique”
as one that was previously selected. Id.
For the reasons above, we determine that “using a selected known
coding technique to generate coded object pixel information,” does not
require a separate step of ‘selecting’ from among available system options,
as argued by Patent Owner.
C. Description of Prior Art References
Petitioner’s challenge primarily relies on Talluri and Monro. See
Pet. 6. The remaining references (Lin and Chai) are relied on as additional
secondary references to address certain specific limitations in the claims.
1. Talluri (Exhibit 1003)
Talluri is an article titled “A Robust, Scalable, Object-Based Video
Compression Technique for Very Low Bit-Rate Coding.” Ex. 1003, 221.
Talluri describes an encoding scheme for MPEG-4. Id. at Abstract, 222.
Talluri describes an object-based video compression technique for very low
bitrate coding of a video stream that “achieves efficient compression by
separating coherently moving objects from stationary background and
compactly representing their shape, motion, and the content.” Id. at 221.
IPR2020-00044
Patent 6,584,229 B1
19
Talluri discloses a five-step encoding process. The first step—
“moving object detection”—involves “segment[ing] the images in the video
sequence into moving objects and stationary background.” Id. The
background and foreground objects are then tiled with macroblocks so as to
generate macroblock-based regions. Id. at 225 (tiling of moving object).
The second step—“shape representation”—involves coding the shape of the
object region. Id. at 225. Talluri describes a “bitmap mask” which uses bits
(0s and 1s) to indicate whether or not macroblocks are part of the region. Id.
In the third step, “motion estimation is used to predict the contents of
the segmented regions from the previous images.” Id. at 226. “[T]he goal
[of this step] is to remove temporal redundancies of the video signal.” Id.
The fourth and the fifth steps apply additional processing to portions of the
image that are not efficiently coded using motion compensation alone (e.g.,
“areas of rapid change”). Id. at 226. Talluri discloses encoding residual
signals using a “DCT approach” that involves a DCT transform, and then
scalar quantization, and variable length coding of the DCT coefficients using
run-length and Huffman coding. Id.
2. Lin (Ex. 1006)
Lin is a patent titled “Method For Determining To Skip Macroblocks
In Encoding Video.” Ex. 1006, code (54). Lin’s “Background” section
explains that during encoding, certain macroblocks can be skipped:
According to the MPEG-2 standard, it is not strictly necessary to
encode each macroblock. Rather, if a prediction error
macroblock has a zero spatial offset motion vector, and only zero
valued quantized block data, no data need be provided in the
bitstream for the macroblock. Instead, indicator information is
placed in the bitstream to indicate that a macroblock has been
skipped.

Id. at 1:41–48.
IPR2020-00044
Patent 6,584,229 B1
20
In determining whether to skip or encode a macroblock, Lin discloses
that, for a given macroblock in the current frame, its luminance values are
compared to macroblocks in the reference picture. Id. at 4:54–59. If the
best match is in the same location in the reference picture, a check is
performed to determine if the luminance difference between the blocks is
below a first threshold. Id. at 5:24–28.
This would indicate that the luminance of the current macroblock
of the current picture is nearly identical to the luminance of the
prediction macroblock in the previous picture at the same spatial
row and column coordinates. In such a case, it is possible that the
current macroblock has not changed from the previous picture.
This typically happens if the macroblock contains a portion of
the image that is static.

Id. at 5:33–39.
Lin further discloses that, if the luminance values sufficiently match, a
similar comparison is performed based on the chrominance data. Id. at
5:48–6:5. “[I]f the chrominance MAD is less than a second threshold, the
processor 12 skips the current macroblock in step 120 by inserting an
indication into the bitstream that indicates that the macroblock is skipped.”
Id. at 6:7–11.
III. ANALYSIS OF THE CHALLENGED CLAIMS
A. Obviousness
A claim is unpatentable under 35 U.S.C. § 103(a) if the differences
between the claimed subject matter and the prior art are such that the subject
matter, as a whole, would have been obvious at the time the invention was
made to a person having ordinary skill in the art to which said subject matter
pertains. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406 (2007). The
question of obviousness is resolved on the basis of underlying factual
determinations, including: (1) the scope and content of the prior art; (2) any
IPR2020-00044
Patent 6,584,229 B1
21
differences between the claimed subject matter and the prior art; (3) the level
of skill in the art; and (4) where in evidence, so-called “secondary
considerations” may be found, including commercial success, long-felt but
unsolved needs, failure of others, and unexpected results. Graham v. John
Deere Co., 383 U.S. 1, 17–18 (1966) (“the Graham factors”).
B. Challenge to Claims 1, 3, 5, and 6 based on Talluri and Lin
Petitioner contends claims 1, 3, 5, and 6 would have been obvious
over the combination of Talluri and Lin. Pet. 28–54. Petitioner supports
this assertion with testimony from its expert, Dr. Saber. Ex. 1002 ¶¶ 94–
154.
Petitioner begins its analysis by arguing that a person of ordinary skill
would have been motivated to combine Talluri and Lin. Pet. 31–34. Talluri
and Lin relate to video coding techniques for use with MPEG video
encoding standards, i.e. Talluri relates to MPEG-4 and Lin relates to MPEG-
1 and MPEG-2. Pet. 31–32 (citing Ex. 1003, Abstract, 223; Ex. 1006, 3:15–
19; Ex. 1013 [unasserted reference], 77 (noting MPEG-4 would “include
most technical features of the prior video and still-picture coding
standards”)).
Petitioner argues, based on disclosures in Talluri and Lin and based on
a person of ordinary skill in the art’s general familiarity with skipping
blocks, that “a [person of ordinary skill in the art] would have found it
obvious to use Lin’s technique for determining when to skip macroblocks to
efficiently encode Talluri’s ‘stationary background’ regions.” Pet. 33 (citing
Ex. 1002 ¶ 107). Additionally, Petitioner asserts that Talluri “does not
describe all the steps involved in performing motion estimation and motion
compensation,” and suggests a person of ordinary skill in the art would have
been motivated to look to Lin’s description of skipping macroblocks for
IPR2020-00044
Patent 6,584,229 B1
22
those details to “achieve greater efficiency” and “reduce computation.” Id.
at 33–34 (citing Ex. 1003, 226; Ex. 1006, 3:15–19; Ex. 1002 ¶¶ 107–108).
Moreover, Petitioner demonstrates sufficiently that the skilled artisan
would have had a reasonable expectation of success in making the
combination. Id. at 34. Petitioner’s declarant states, “Lin notes that its
method is generally applicable to standard, block-based coders.” Ex. 1002
¶ 108.
Patent Owner does not argue against the combination. See generally
PO Resp. In summary, based on the information set forth in the Petition and
the testimony of Dr. Saber, we are persuaded that Petitioner has
demonstrated sufficiently both the motivation to combine these references
and the reasonable expectation of success for an ordinary skilled artisan.
1. Claim 1
Petitioner’s element-by-element analysis of independent claim 1 in
relation to Talluri and Lin appears at pages 34–50 of the Petition. Petitioner
supports its analysis with testimony from Dr. Saber. Ex. 1002 ¶¶ 109–145.
Petitioner demonstrates that each element of claim 1 is disclosed or
suggested by the combination of Talluri and Lin.
a. Preamble
“[a] method, for use in an macroblock-based object-oriented
coding of a[n] image signal, wherein the image signal has a
stationary background region and an object region and contains
a current frame and a previous frame, each frame including a
plurality of macroblocks, comprising the steps of,”

Petitioner identifies the claim term “[a] method, for use in an
macroblock-based object-oriented coding of a[n] image signal, wherein the
image signal has a stationary background region and an object region and
contains a current frame and a previous frame, each frame including a
IPR2020-00044
Patent 6,584,229 B1
23
plurality of macroblocks, comprising the steps of,” as recited in the preamble
of claim 1 in Talluri’s teaching of macroblocks that are in the MPEG-4
frames. Pet. 34 (citing Ex. 1003, 223; Ex. 1002 ¶ 113). Petitioner also
identifies Talluri’s moving object detection step that “segment[s] the images
in the video sequence into moving objects and stationary background.” Id.
at 35 (citing Ex. 1003, 224). Petitioner also identifies Talluri’s image
signals that include a current frame and a previous frame. Pet. 35 (citing Ex.
1003, 224 (“the absolute difference of the current and previous images”));
Ex. 1002 ¶¶ 115–116).
We have reviewed the record, and find that Petitioner provides
sufficient evidence that the prior art teaches the preamble of claim 1.4 Patent
Owner makes no arguments contesting the prior art teachings of these claim
limitations. See generally, PO Resp.
b. step a)
“dividing the stationary background region and the object
region from an inputted video in a macroblock-by-macroblock
basis by using a difference between the previous frame and the
current frame”

Petitioner identifies step a)—the claim term “dividing the stationary
background region and the object region from an inputted video in a
macroblock-by-macroblock basis by using a difference between the previous
frame and the current frame”—recited in claim 1 in Talluri’s motion
segmentation block of Figure 2 (not shown). Id. at 36–37 (citing Ex. 1003,
224, Fig. 2; Ex. 1002 ¶¶ 117–119). For example, Petitioner also identifies

4 We need not determine whether the preamble of claim 1 is limiting because
Petitioner has shown Talluri discloses the preamble. See Nidec, 868 F.3d at
1017.
IPR2020-00044
Patent 6,584,229 B1
24
step a) in Talluri’s calculating and tiling of rectangles with macroblocks in
the regions in the frames and background. Id. at 36 (citing Ex. 1003, 225,
226, 231). Petitioner also identifies step a) in Talluri’s “thresholding the
absolute difference of the current and previous images.” Id. at 37 (citing Ex.
1003, 224, 225; Ex. 1002 ¶¶ 117–119) (emphasis omitted).
Petitioner admits that Talluri performs a two-step process to divide the
foreground and background region. Reply 7. Specifically,
Talluri discloses a two-step process that is used to divide a video
into background and object regions composed of macroblocks.
First, “a segmentation mask is generated for the current image
using image differencing, morphology, and connected
components analysis.” Talluri at p. 224. This step identifies
image regions detected as moving objects. Id.; id. at p. 225, Fig.
4(a). Next, a bounding rectangle is calculated for each moving
object region and the rectangle is “tile[d] … with16 X 16-pixel
macroblocks.” Id. at p. 225. This step generates macroblock-
based regions from the earlier-detected image regions. Ex. 1027
(Saber) ¶ 212.

Id. Petitioner contends, however, that a person of ordinary skill “would
readily appreciate that Talluri’s moving object detection and macroblock
refinement steps operate together to divide a video into macroblock-based
moving object and stationary background regions.” Id. Patent Owner
argues Talluri’s segmentation step, which is done over the entire background
and foreground region, cannot meet the claim limitation under its
construction of the claim language. Id. at 12. As explained above, we did
not adopt Patent Owner’s construction of the “dividing” step; thus, this
argument is not commensurate with the proper scope of the claims.
As noted above, Petitioner explains “[n]othing in the language of the
‘dividing’ step mandates a single step approach,” and “[a]s long as the
dividing is done on a macroblock-by-macroblock basis and ‘a difference
IPR2020-00044
Patent 6,584,229 B1
25
between the previous frame and the current frame’ is used, the limitation is
satisfied. Reply 8. We are persuaded by this contention, which is consistent
with our construction and with the claim language.
Patent Owner also argues that Talluri does not disclose dividing the
background in a macroblock-by-macroblock basis. Sur-Reply 6; Tr. 35:6–
19. Specifically, Patent Owner argues that Talluri
discloses, the second step ‘calculates a bounding rectangle for
each of the moving object regions and tiles the rectangles with
16 16-pixel macroblocks’. [Ex. 1003] at 225 (emphasis added).
In other words, in Talluri, the background is never divided or
defined in a ‘macroblock-by-macroblock basis’. Talluri only
divides the ‘bounding rectangle’ of ‘the moving object regions’
with macroblocks.

Sur-Reply 6. Petitioner contends that this is a new argument raised for the
first time in the Sur-Reply. Tr. 13:12–17. We agree that this was newly
raised in the Sur-Reply because it does not respond to any new argument in
Petitioner’s Reply. Nevertheless, we also determine that Petitioner has
shown sufficiently that Talluri teaches this limitation. Specifically,
Petitioner’s declarant asserts, “The same technique [calculating a bounding
rectangle] is applied to the background object.” Ex. 1002 ¶ 118 (citing Ex.
1003 [Talluri], 231 (noting that the shape of the background object is coded
using either the block-based or spline-based technique); see also Pet. 36
(stating the same thing). At the hearing, Patent Owner suggested that this
disclosure in Talluri was “separated by many, many, many pages from their
segmenting and calculating tiling of the moving object 15 region [and may
not be] related at all.” Tr. 35:6–19. We disagree. We credit Petitioner’s
declarant and agree that Talluri discloses defining the background region “in
a ‘macroblock-by-macroblock basis.”
IPR2020-00044
Patent 6,584,229 B1
26
For the reasons above, we determine that Talluri’s two-step process of
segmentation of the background and foreground regions followed by tiling
into macroblocks reads on the limitation of “dividing the stationary
background region and the object region from an inputted video in a
macroblock-by-macroblock basis by using a difference between the previous
frame and the current frame.”
c. Step b)
“coding shape information of the object region by using a known
coding technique to generate coded shape information.”
Petitioner relies on Talluri to meet the step b) limitation of “coding
shape information of the object region by using a known coding technique to
generate coded shape information.” Pet. 37–39. For example, Petitioner
relies on Talluri’s disclosure of “a macroblock-based shape coding technique
which generates ‘bitmap masks’ of the segmented regions of the video.”
Id. at 38 (citing Ex. 1003, 225).
d. Step c)
“coding pixel information of each macroblock contained in the
object region by using a selected known coding technique to
generate coded object pixel information.”
Petitioner relies on Talluri to meet the step c) limitation of “coding
pixel information of each macroblock contained in the object region by
using a selected known coding technique to generate coded object pixel
information.” Id. at 39–41. Referring to Talluri’s DCT approach and
production of motion vectors and coded residual data produced by encoding
the object region, Petitioner asserts a person of ordinary skill at the time of
the invention would have understood that this information is “coded object
pixel information.” Id. at 39–40 (citing Ex. 1003, 224, 226, Fig. 2; Ex. 1002
¶ 127).
IPR2020-00044
Patent 6,584,229 B1
27
As explained above, Patent Owner argues that the “Petition makes no
mention of the ‘selected’ qualifier apart from its reproduction of claim 1
(Pet. 1) and the inclusion of this term in certain section headings.” PO Resp.
10. We do not adopt Patent Owner’s proposed construction of “selected” as
making a “selection from available system options.” Thus, the DCT
approach cited by the Petition is the “selected” known coding technique used
by Talluri, consistent with the claim limitation. Pet. 39–40; Ex. 1002 ¶ 127
(“the ’229 patent describes using a conventional encoding approach that
involves a DCT transformation, quantization, and variable length coding.
’229 patent at 3:24-29. Talluri’s ‘DCT approach’ involves the same 5
steps”).
e. Step d)
“generating coded pixel information of a previous frame
macroblock corresponding to each current frame macroblock
contained in the stationary background region as coded
stationary pixel information.”
Petitioner relies on the combination of Talluri and Lin to meet the step
d) limitation of “generating coded pixel information of a previous frame
macroblock corresponding to each current frame macroblock contained in
the stationary background region as coded stationary pixel information.” Id.
at 41–44. Petitioner relies on two alternative contentions based on
Petitioners two proposed constructions. As to the first construction that
requires encoding the previous frame, Petitioner relies on the combination of
Talluri and Lin. For example, Petitioner relies on Talluri’s disclosure “that
the previous frame is encoded and then decoded to make the reconstructed
frame used for motion compensation.” Id. at 43 (citing Ex. 1003, 231
(“After all of the objects in the first frame have been coded, a reconstructed
IPR2020-00044
Patent 6,584,229 B1
28
frame is made, combining decoded images of the background and each
object into one frame.”)). Petitioner further relies on Lin’s disclosure that
the reference picture is preferably encoded and then decoded
(i.e., reconstructed). [Ex. 1006, 4:60–63] (‘Preferably, the
reference pictures have been previously encoded, decoded and
stored in a picture buffer of the memory 14. This is so that
predictions are formed using the same decoded pictures available
at the decoder.’).

Id. at 43 (emphasis omitted).
As to the second construction that requires encoding the current
frame, Petitioner relies on the combination of Talluri and Lin. For example,
Petitioner relies on Talluri’s disclosure that “motion estimation is used to
predict the contents of segmented regions [of the current image] from the
previous images.” Id. at 41–42 (citing Ex. 1003, 226). Petitioner further
relies on Lin’s “motion compensation technique whereby the current frame
is encoded using coded macroblocks from a previous picture.” Id. at 42
(citing Ex. 1006, 1:18–40; 4:5–7 (“The reference picture for a P picture is
the immediately preceding I or P picture in the sequence of pictures.”)).
f. Step e)
“storing or transmitting coded data coded shape information,
coded object pixel information and coded stationary pixel
information as coded image signal”

Petitioner asserts that Talluri discloses step e)—“storing or
transmitting coded data coded shape information, coded object pixel
information and coded stationary pixel information as coded image
signal”—as recited in claim 1. Id. at 45–46. For example, as to “coded
shape data,” Petitioner relies on Talluri’s disclosure that “as part of
transmitting the coded bitstream, coded shape information is transmitted:
IPR2020-00044
Patent 6,584,229 B1
29
‘[i]n transmitting a coded shape [to the decoder], the encoder first sends a 3-
b tag identifying the format of the shape data to follow. The decoder reads
this tag and decodes the shape data appropriately.’” Id. at 46 (quoting
Ex. 1003, 225–226, 229; Ex. 1002 ¶ 136). As to coded object pixel data and
coded stationary pixel data, Petitioner relies on the same disclosure as it
relies on for “coding pixel information” discussed in regard to step c),
above. Id. at 46 (citing Ex. 1003, 224, 231, Fig. 2; Ex. 1002 ¶¶ 137–38)).
g. Wherein Clause
“wherein the step d) includes the step of reusing corresponding
coded pixel information macroblock contained in the previous
frame without coding the pixel information of each macroblock
contained in the current frame when a difference between a pixel
value of the macroblock of the current frame and that of the
macroblock of the previous frame in the same position is
identical to or smaller than a predetermined threshold value.”

Petitioner asserts that Talluri in combination with Lin discloses the
additional limitation “wherein the step d) includes the step of reusing
corresponding coded pixel information macroblock contained in the
previous frame without coding the pixel information of each macroblock
contained in the current frame when a difference between a pixel value of
the macroblock of the current frame and that of the macroblock of the
previous frame in the same position is identical to or smaller than a
predetermined threshold value,” as recited in claim 1. Id. at 47–50. For
example, Petitioner relies on Lin’s disclosure that in comparing the current
block with a previous block “if there is ‘zero spatial offset’ with the best
matching block in the previous frame, and the luminance difference falls
below a first threshold, and the chrominance difference falls below a second
threshold, then a block is coded as a skipped macroblock.” Id. at 47–48
IPR2020-00044
Patent 6,584,229 B1
30
(citing Ex. 1006, 5:24–42, 5:38–39, 5:48–6:7; Ex. 1002 ¶¶ 140–41). In
particular, “zero spatial offset” means that “it is in the same location as the
current block in the current frame.” Id. (quoting Ex. 1006, 5:24–42, citing
id., 5:38–39).
Patent Owner argues that “Lin relies on comparing the current
macroblock against a previous macroblock at an offset relative to the current
macroblock.” PO Resp. 22. Patent Owner is correct that Lin compares with
macroblocks “at increments of 1/2 pixel row and column offsets and
compares them to the current macroblock.” Ex. 1006, 4:65–66 (emphasis
added). Nevertheless, as explained above, Lin specifically contemplates a
case where there is zero spatial offset. Id. at 5:24–42. Patent Owner
discounts Lin’s use of a zero spatial offset because it is allegedly “only
based on no suitable motion vector being found.” PO Resp. 22. Patent
Owner does not provide a citation to Lin for this proposition. Id. To the
contrary, Lin states, in reference to zero spatial offset, that if
the luminance of the current macroblock of the current picture is
nearly identical to the luminance of the prediction macroblock in
the previous picture at the same spatial row and column
coordinates. In such a case, it is possible that the current
macroblock has not changed from the previous picture.

Ex. 1006, 5:33–38. Patent Owner does not explain sufficiently how the case
in which “no suitable motion vector is found” would prevent Lin from
meeting the claim limitation. In sum, Patent Owner does not explain
sufficiently why selecting a “prediction macroblock” in the same position as
the current macroblock could not meet the claim limitation to “the same
position.”
Additionally, Patent Owner asserts,
IPR2020-00044
Patent 6,584,229 B1
31
it is unclear from the teachings of Lin whether the first threshold
is used to select a certain candidate prediction macroblock from
among multiple candidate prediction macroblocks based on the
MAD calculated for each of the candidate prediction macroblock
or not. What is known is that Lin never teaches comparing any
macroblock from a current picture with another macroblock from
a reference picture in the same position.

PO Resp. 22–23. Again, Patent Owner does not provide a citation to Lin for
this proposition. Id. To the contrary, Lin discloses determining “if the
luminance MAD generated for the prediction macroblock is below a first
threshold.” Ex. 1006, 5:26–28 (emphasis added). Patent Owner has
provided no independent evidence to support its assertions, such as a
citations to Lin to explain certain assertions (e.g., no suitable motion vector
found, or MAD calculated for multiple candidate prediction block not for a
single block), expert testimony, or scientific publications. An argument
made by counsel in a brief does not substitute for evidence lacking in the
record. Estee Lauder, Inc. v. L’Oréal, S.A., 129 F.3d 588, 595 (Fed. Cir.
1997); see also Icon Health & Fitness, Inc. v. Strava, Inc., 849 F.3d 1034,
1043 (Fed. Cir. 2017) (“Attorney argument is not evidence.”).
Additionally, in support of the reusing limitation, Petitioner asserts
that “[a person of ordinary skill in the art] would have known that where
macroblocks are skipped, corresponding macroblocks in the reference
picture would be reused.” Pet. 48 (citing Ex. 1002 ¶ 142). As support for
this assertion, Petitioner relies on the 1993 MPEG-1 standard document that
states, “Any macroblocks thus skipped over are called ‘skipped
macroblocks’. The decoder copies them from the previous picture into the
current picture.” Id. (quoting Ex. 1014, 93). Petitioner admits that in
skipping macroblocks Lin expressly states: “no data [is] provided in the
bitstream” and, “[i]instead, indicator information is placed in the bitstream
IPR2020-00044
Patent 6,584,229 B1
32
to indicate that a macroblock has been skipped.” Pet. 29–30 (quoting Ex.
1006, 1:41‒48).
Patent Owner argues that “when skipping a macroblock, Lin
disclosing providing no pixel data and, instead, placing a new ‘indic[a]tor’
into the bitstream . . . is plainly distinguishable from the ‘reusing’ step.” PO
Resp. 21.5 Petitioner responds that, as stated above, the Petition explains
that a person of ordinary skill would have been well-aware of the skipped
macroblock technique which was the standard way to signal that previous
frame macroblocks should be reused for those portions of the current frame
that have not changed. Reply 10 (citing Pet. 10–11; Ex. 1002 ¶¶ 44–50,
101). Petitioner further contends a person of ordinary skill in the art “would
regard the notion of duplicating redundant data for every unchanged
macroblock between successive frames—Uniloc’s apparent interpretation—
as incredibly inefficient and illogical.” Id. at 10–11 (citing Ex. 1027 ¶ 217).
Petitioner’s declarant also testifies that a person of ordinary skill in the art
would have understood the ’229 patent specification as using the skipped
macroblock technique. Ex. 1027 ¶ 217 (cited in Petition at 11). We agree
with Petitioner that the claim does not recite and is not limited to a
“corresponding coded pixel information macroblock contained in the
previous frame” that must be retransmitted to be reused. Thus, the issue

5 Petitioner argues that “this argument is utterly contrary to Uniloc’s position
in district court where its entire infringement argument for the ‘reusing’
limitation is based on skipped macroblocks.” Reply 9–10 (citing Ex. 1025,
14–15). We considered this evidence and acknowledge this apparent
inconsistency but also note that the position taken by a party in the District
Court is not a final (or even non-final) ruling from the District Court. Thus,
we consider the intrinsic and extrinsic evidence and the arguments made in
this IPR to arrive at the proper construction of claim terms.
IPR2020-00044
Patent 6,584,229 B1
33
before us is whether Lin discloses reusing coded pixel information of a
previous macroblock (whether or not it was retransmitted or simply saved
from a previous encoding).
Petitioner’s ground is based on obviousness, and we credit Petitioner’s
declarant who testifies that one of ordinary skill would have understood that
when a macroblock is skipped (not encoded), the previous macroblock in
that position would be used. Ex. 1002 ¶ 142. Although knowledge of one of
ordinary skill may be properly considered in an obviousness analysis, it
“cannot be used as a wholesale substitute for reasoned analysis and
evidentiary support, especially when dealing with a limitation missing from
the prior art references specified.” Arendi S.A.R.L. v. Apple Inc., 832 F.3d
1355, 1361–62 (Fed. Cir. 2016). In recognizing the role of common
knowledge, “[the Federal Circuit has] emphasized the importance of a
factual foundation to support a party’s claim about what one of ordinary skill
in the relevant art would have known.” Randall Mfg. v. Rea, 733 F.3d 1355,
1362 (Fed. Cir. 2013) (emphasis added)); see also Koninklijke Philips N.V.
v. Google LLC, 948 F.3d 1330, 1338 (Fed. Cir. 2020) (finding a limitation at
issue was not missing from the prior art, stating “here the Board relied
on expert evidence, which was corroborated by Hua, in concluding that
pipelining was not only in the prior art, but also within the general
knowledge of a skilled artisan.”).
Here, Petitioner’s declarant states that the allegedly missing limitation
would have been known to one of ordinary skill and provides a reasoned
explanation of the basis of that knowledge, which is supported by an
underlying reference. Ex. 1002 ¶ 142 (quoting Ex. 1014, 93); see also Ex.
1002 ¶ 28 (MPEG-1 standard was published in 1993). In this IPR, Petitioner
has asserted that Talluri uses the previously coded macroblock, shown that
IPR2020-00044
Patent 6,584,229 B1
34
an industry standard explains how a macroblock is reused, and has reasoned,
based on its declarant’s testimony, that an ordinarily skilled artisan would
have considered the alternative to be “incredibly inefficient and illogical.”
Reply 10–11 (citing Pet. 10–11; Ex. 1002 ¶¶ 44–50, 101; Ex. 1027 ¶ 217).
Additionally, Patent Owner did not respond to this argument; instead
Patent Owner suggested that Petitioner must rely on inherency. Sur-Reply
16. Specifically, Patent Owner argues in its Sur-Reply that “Petitioner fails
to meet the exacting standard to prove inherency.” Sur-Reply 16. Because
this ground is an obviousness ground and Petitioner has not indicated that it
relies on inherency, we do not determine inherency is relevant to this
issue. “[W]hat would have been obvious to a skilled artisan is not a valid
consideration in an inherency analysis.” See W.L. Gore & Assocs., Inc. v.
Garlock, Inc., 721 F.2d 1540, 1555 (Fed. Cir. 1983) (“Inherency and
obviousness are distinct concepts.”) (citing In re Spormann, 363 F.2d 444,
448 (CCPA 1966)).
h. Conclusion
Based on the information set forth in the Petition and the testimony of
Dr. Saber, we are persuaded Petitioner has shown by a preponderance of the
evidence that claim 1 would have been obvious over Talluri and Lin.6

2. Claim 3
This claim depends from claim 1 and further recites “wherein the step
c) includes the steps of coding pixel information of each macroblock
contained in the current frame when a difference between a pixel value of

6 Neither party has argued that secondary considerations or objective
evidence of nonobviousness exists. Thus, we do not address secondary
considerations or objective evidence of nonobviousness.
IPR2020-00044
Patent 6,584,229 B1
35
the macroblock of the current frame and that of the macroblock of the
pr[e]vious frame in the same position is bigger than a predetermined
threshold value.” Pet. 50. Petitioner asserts that the combination of Talluri
and Lin discloses a motion compensated prediction technique that
determines when macroblocks can be skipped. Id. Petitioner asserts Lin
discloses this limitation because Lin compares the luminance of a
macroblock in the current frame with a macro block in a reference frame and
if the luminance is greater than or equal to a threshold the macroblock gets
encoded. Id. at 50–51 (citing Ex. 1006, 5:25–28, 5:45–48, 6:25–50;
Ex. 1002 ¶¶ 146–49).
Patent Owner’s Response does not provide a rebuttal to Petitioner’s
analysis of this dependent claim in light of the teachings of Talluri and Lin.
Based on the information set forth in the Petition and the testimony of Dr.
Saber, we are persuaded Petitioner has shown by a preponderance of the
evidence that claim 3 would have been obvious over Talluri and Lin.
3. Claim 5
This claim depends from claim 1 and further recites “wherein the
known coding technique includes a discrete cosine transformation, a
quantum-transformation, and a coefficient variable length transformation.”
Pet. 51. Petitioner asserts that Talluri discloses this limitation because
Talluri uses Huffman coding which a person of ordinary skill at the time of
the invention would know, according to Petitioner, is a type of variable
length coding. Id. at 52 (citing Ex. 1002 ¶ 151; Ex. 1012 [an unasserted
reference describing variable length Huffman codes in MPEG coding], 29);
Ex. 1006 [Lin–for “corroboration”], 1:34–40.
Patent Owner’s Response does not provide a rebuttal to Petitioner’s
analysis of this dependent claim in light of the teachings of Talluri and Lin.
IPR2020-00044
Patent 6,584,229 B1
36
Based on the information set forth in the Petition and the testimony of Dr.
Saber, we are persuaded Petitioner has shown by a preponderance of the
evidence that claim 5 would have been obvious over Talluri and Lin.
4. Claim 6
This claim depends from claim 1 and further recites “wherein the
storage and transmission are a computer readable media and communication
network.” Pet. 53. Petitioner asserts that Talluri’s communication channels
and multimedia database meet this limitation. Id. (citing Ex. 1003, 222;
Ex. 1006 [Lin – for “corroboration”], 3:35–38; Ex. 1002 ¶ 152).
Patent Owner’s Response does not provide a rebuttal to Petitioner’s
analysis of this dependent claim in light of the teachings of Talluri and Lin.
Based on the information set forth in the Petition and the testimony of Dr.
Saber, we are persuaded Petitioner has shown by a preponderance of the
evidence that claim 6 would have been obvious over Talluri and Lin.

C. Challenge to Claims 1, 5, and 6 based on Monro and Chai
Petitioner contends these claims would have been obvious over the
combination of Monro and Chai. Pet. 54–73. Petitioner supports this
assertion with testimony from Dr. Saber. Ex. 1002 ¶¶ 160–203.
For the reasons discussed above, Petitioner has shown, by a preponderance
of the evidence, that claims 1, 3, 5, and 6 are unpatentable as obvious over
Talluri and Lin. In addressing this ground, we have addressed all of the
challenged claims. See 35 U.S.C. § 318(a) (requiring the Board to “issue a
final written decision with respect to the patentability of any patent claim
challenged by the petitioner and any new claim added under section
316(d)”); see also SAS Inst. Inc. v. Iancu, 138 S. Ct. 1348, 1359 (2018)
(holding that a petitioner “is entitled to a final written decision addressing all
IPR2020-00044
Patent 6,584,229 B1
37
of the claims it has challenged”). Accordingly, we need not and do not
decide whether Petitioner has shown by a preponderance of the evidence that
claims 1, 5, and 6 are unpatentable as obvious over Monro and Chai. Cf. In
re Gleave, 560 F.3d 1331, 1338 (Fed. Cir. 2009) (not reaching other grounds
of unpatentability after affirming the anticipation ground); see also Boston
Sci. Scimed, Inc. v. Cook Grp. Inc., 809 F. App’x 984, 990 (Fed. Cir. Apr.
30, 2020) (non-precedential) (recognizing that the “Board need not address
issues that are not necessary to the resolution of the proceeding” and, thus,
agreeing that the Board has “discretion to decline to decide additional
instituted grounds once the petitioner has prevailed on all its challenged
claims”).
IV. CONCLUSION
For the foregoing reasons, we determine Petitioner has established by
a preponderance of the evidence the unpatentability of claims 1, 3, 5 and 6
of the ’229 patent.7

7 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
IPR2020-00044
Patent 6,584,229 B1
38
In summary:
Claim(s) 35 U.S.C. §8 References/ Basis
Claim(s)
Shown
Unpatentable
Claim(s) Not
Shown
Unpatentable
1, 3, 5, 6 103(a) Talluri, Lin 1, 3, 5, 6
1, 5, 6 103(a) Monro, Chai
Overall
Outcome
1, 3, 5, 6

V. ORDER
Accordingly, it is
ORDERED that claims 1, 3, 5, and 6 of U.S. Patent No. 6,584,229 B1
are unpatentable; and
FURTHER ORDERED that parties to the proceeding seeking judicial
review of this Final Written Decision must comply with the notice and
service requirements of 37 C.F.R. § 90.2.

8 As discussed in Section III.B above, we do not reach the alternate ground
raised for claims 1, 5, and 6, as we have determined those claims are
unpatentable based on first ground set forth in this table.
IPR2020-00044
Patent 6,584,229 B1
39
FOR PETITIONER:
Babak Tehranchi
Tehranchi-ptab@perkinscoie.com

Patrick McKeever
Mckeever-ptab@perkinscoie.com

FOR PATENT OWNER:

Jeffrey Huang
jeff@etheridgelaw.com

Jeffrey Stephens
jstephens@etheridgelaw.com

Brett Mangrum
brett@etheridgelaw.com

James Etheridge
jim@etheridgelaw.com

Ryan Loveless
ryan@etheridgelaw.com