BLACKBERRY LIMITED

Patent Trials and Appeals Board

Dec 22, 2020

2019004665 (P.T.A.B. Dec. 22, 2020)

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.
14/277,172 05/14/2014 Dake HE 101-0216USP1 6959
14551 7590 12/22/2020
Rowand LLP (BlackBerry)
3080 Yonge Street, Suite 6060
Toronto, ONTARIO M4N 3N1
CANADA
EXAMINER
MAHMUD, FARHAN
ART UNIT PAPER NUMBER
2483
NOTIFICATION DATE DELIVERY MODE
12/22/2020 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):
mailbox@rowandlaw.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte DAKE HE, XIAOFENG WANG, JIM WANG,
TIANYING JI, and DAVID FLYNN
Appeal 2019-004665
Application 14/277,172
Technology Center 2400

Before JAMES B. ARPIN, DAVID J. CUTITTA II, and
MICHAEL J. ENGLE, Administrative Patent Judges.

CUTITTA, Administrative Patent Judge.

DECISION ON APPEAL
STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1, 2, 4–13, and 15–27, all of the claims
under consideration.2 We have jurisdiction under 35 U.S.C. § 6(b).
We AFFIRM.

1 We use the word “Appellant” to refer to “applicant” as defined in 37
C.F.R. § 1.42(a). Appellant identifies the real party in interest as BlackBerry
Limited. Appeal Br. 2.
2 Appellant cancelled claims 3 and 14. Appeal Br. 27, 29.
Appeal 2019-004665
Application 14/277,172

2
CLAIMED SUBJECT MATTER
Invention
Appellant’s claimed subject matter relates to “context-adaptive video
coding.” Spec. ¶ 1.3 In particular, Appellant explains that a “context
initialization buffer 108 [is used to store] two or more context model states
120 corresponding to [a] context model state following context-adaptive
entropy encoding of a respective two or more previous slices or pictures.”
Id. ¶ 48. An “entropy encoder [then] selects one of the stored context model
states from the buffer” and initializes the context model “using the context
model state selected from the buffer.” Id. ¶¶ 62, 63. This may provide
sufficient “values to adapt the probabilities associated with a particular set of
contexts quickly enough,” even if data is sparse. Id. ¶ 3.
Exemplary Claim
Claims 1, 12, 23, 24, 25, 26, and 27 are independent. Claim 1,
reproduced below with limitations at issue italicized, exemplifies the
claimed subject matter:
1. A method of encoding video using a video
encoder, the video encoder employing context-adaptive entropy
encoding using a context model having a plurality of contexts,
the context model having a context model state defining the
respective probability associated with each context defined in
the context model, the video encoder storing a predefined
context model state for initialization of the probabilities of the
context model, and the video encoder including a buffer, the
method comprising:

3 We refer to: (1) the originally filed Specification filed May 14, 2014
(“Spec.”); (2) the Final Office Action mailed July 12, 2018 (“Final Act.”);
(3) the Appeal Brief filed December 3, 2018 (“Appeal Br.”); and (4) the
Examiner’s Answer mailed March 22, 2019 (“Ans.”).
Appeal 2019-004665
Application 14/277,172

3
context-adaptively encoding a first picture, including
progressively updating the contexts of a context model state as
bins are coded, and storing, in the buffer,
a first context model state associated with the first picture
after encoding of the first picture;
context-adaptively encoding a second picture, including
progressively updating the context model state as bins are
coded and storing, in the buffer, a second
context model state associated with the second picture
after encoding of the second picture;
for encoding a current picture of the video, wherein the
first and second pictures are previously-encoded pictures of the
video and wherein the first context model state and the second
context model state are at least two different stored context
model states in the buffer for the context model,
selecting one of the at least two stored context model
states from the buffer;
initializing the context model for context-adaptively
encoding the current picture using the selected one of the at
least two stored context model states; and
context-adaptively entropy encoding the current picture
to produce a bitstream of encoded data; and
outputting the bitstream of encoded data.
Appeal Br. 26 (Claims Appendix).
REFERENCES AND REJECTION
The Examiner rejects claims 1, 2, 4–13, and 15–27 under 35 U.S.C.
§ 103 as obvious over the combined teachings of He et al. (US
2012/0014457 A1, published Jan. 19, 2012) (“He”) and Cote et al. (US
2015/0092834 A1, published Apr. 2, 2015) (“Cote”). Final Act. 10–17.4

4 The Examiner rejected claims 1, 2, 4–13, and 15–27 under 35 U.S.C.
§ 101, as directed to patent-ineligible subject matter without significantly
Appeal 2019-004665
Application 14/277,172

4
OPINION
We review the appealed rejection for error based upon the issues
identified by Appellant and in light of Appellant’s arguments and evidence.
Ex parte Frye, 94 USPQ2d 1072, 1075 (BPAI 2010) (precedential).
Arguments not made are waived. See 37 C.F.R. § 41.37(c)(1)(iv) (2018).
Appellant does not persuade us that the Examiner errs, and we adopt as our
own the findings and reasons set forth by the Examiner to the extent
consistent with our analysis herein. Final Act. 10–12; Ans. 4–10. We add
the following primarily for emphasis.
“selecting one of the at least two stored context model states”
The Examiner finds He teaches or suggests, “selecting one of the at
least two stored context model states from the buffer,” as recited in claim 1.
Final Act. 10 (citing He ¶¶ 82, 83); Ans. 4–5 (citing He ¶¶ 43, 82, and 83).
Of particular relevance, the Examiner relies on He’s discussion of
H.264/AVC and, specifically, of encoding a bit in the jth position based on
the probability of any selected bit in the jth position in previous iterations (i–
1, i–2, etc.) of a given sequence i. Final Act. 10; see He ¶ 82 (“for a bit in
the jth position in a given sequence i, its probability state is selected from
amongst the 64 possible probabilities based on the history of bits in the jth
position in previous sequences (i-1, etc.).”).
Appellant argues that “He’s reference to selecting one of the
probabilities for a (singular) context cannot be equated with selecting a full
stored context model state (i.e.[,] the probabilities associated with every

more. Final Act. 2–4, 7–8. The Examiner has withdrawn that rejection.
Ans. 3.
Appeal 2019-004665
Application 14/277,172

5
context in the model)” because “He is simply describing the act of updating
the probability value associated with a specific context in the course of
coding.” Appeal Br. 20.
The Examiner responds:
Appellant further argues that the probabilities described are not
equated to context model states. However, Paragraph 82, very
clearly states “[i]n the H.264/AVC example the probability
(sometimes termed “state”, “probability state” or “context state”)
for a bit is determined based on its context.” Thus it is clearly
taught that reference frames are held in a buffer, and that context
model states, are referred to and selected from among the 64
possible probabilities based on the history of bits in that
particular position in previous sequences.
Ans. 4–5.
Appellant’s arguments are unpersuasive. Claim 1 recites that “a
context model state defin[es] the respective probability associated with each
context defined in the context model.” Appeal Br. 26 (emphasis added).
Thus, the claim itself indicates that a context model state includes a
probability. In addition, the Examiner finds that “context model states, as
commonly understood in the relevant arts, are referred to and selected from
among the 64 possible probabilities based on the history of bits in that
particular position in previous sequences.” Final Act. 5. Appellant also fails
to persuade us that the Examiner’s interpretation of “context model state” is
inconsistent with Appellant’s Specification. In re Am. Acad. of Sci. Tech
Ctr., 367 F.3d 1359, 1364 (Fed. Cir. 2004). Appellant has not shown that
the Specification defines expressly the term “context model state” as recited
in claim 1. Rather, similar to the claim itself, the Specification describes a
“context model state” as including a probability. Spec. ¶ 41.
Appeal 2019-004665
Application 14/277,172

6
He, in turn, defines a probability associated with each context for bits
in various positions in a given sequence i. For example, He discloses that
“for a bit in the jth position in a given sequence i, its probability state is
selected from amongst the 64 possible probabilities based on the history of
bits in the jth position in previous sequences (i-1, etc.).” He ¶ 82. Similarly,
He teaches selecting a probability state for a bit in the jth position in the
previous sequence i–2. We agree with the Examiner that He’s discussion of
defining a probability for a bit position based on a bit position in a previous
iteration of the sequence (e.g., i–2, i–1, etc.) teaches “selecting one of the at
least two stored context model states,” as recited in claim 1. Accordingly,
Appellant’s argument that He’s probabilities fails to teach the claimed
“context model states” is unpersuasive because Appellant fails to establish
that the Examiner’s interpretation of “context model state,” as recited in
claim 1, is unreasonably broad. See Ans. 5.
Appellant further argues that “He’s description of there being 64 fixed
prescribed probabilities values cannot be equated with storing a snapshot of
all the values of the contexts at a point in time and later selecting one of the
stored snapshots from among the stored snapshots (context model states).”
Appeal Br. 20. This argument is unpersuasive because the argued feature
“storing a snapshot of all the values of the contexts at a point in time” (id.) is
not recited in the claim and, therefore, is not commensurate with the scope
of claim 1.
In addition, the Examiner, in the Answer, finds for the first time that
Cote “very clearly teaches the caching of and selecting from multiple
context states in a context cache, and also the possibility of transitioning to
multiple other states from each context state that is stored.” Ans. 7 (citing
Appeal 2019-004665
Application 14/277,172

7
Cote ¶¶ 115, 116, 146, Figs. 13, 16). Appellant does not rebut the
Examiner’s additional reasoning and findings. Consequently, Appellant
does not persuade us of error in these additional factual findings or in the
Examiner’s conclusion of obviousness.
Accordingly, Appellant fails to show reversible error in the
Examiner’s finding that the cited references teach or suggest “selecting one
of the at least two stored context model states from the buffer,” as recited in
claim 1.
“initializing the context model”
The Examiner finds He teaches or suggests, “initializing the context
model for context-adaptively encoding the current picture using the selected
one of the at least two stored context model states,” as recited in claim 1.
Final Act. 10 (citing He ¶¶ 82, 83, 129–133). Of particular relevance, the
Examiner relies on He’s initialization to zero of indexes i and j. He ¶ 130.
Appellant argues that:
Even if the position j is equated with a “context”, there is no
suggestion in this portion of He that the context state (i.e.[,]
associated probability) is being initialized to a starting value.
Rather, He is simply saying his algorithmic process will start
with sequence 0 and position 0 by initializing the indices to zero.
Appeal Br. 22–23.
Appellant’s argument is unpersuasive because it is not commensurate
with the scope of the claim. Appellant argues that He fails to suggest “that
the context state (i.e.[,] associated probability) is being initialized to a
starting value.” Appeal Br. 23. Claim 1, however, does not recite
initializing the context state, but rather recites “initializing the context
model.” Id. at 26 (emphasis added). Appellant, therefore, does not persuade
Appeal 2019-004665
Application 14/277,172

8
us of error in the Examiner’s finding that He’s “initialization of the indices
to zero is considered an initialization of the context model.” Ans. 22–23.
In addition, the Examiner finds that “Cote very clearly and
unambiguously teaches initializing context models in the context adaptive
encoding.” Ans. 10 (citing Cote ¶¶ 115, 116, 146, Figs. 13, 16). Of
particular relevance, the Examiner finds that Cote’s “method may include
initializing the context model states in a context memory lookup table
(LUT).” Ans. 9 (emphasis omitted) (citing Cote Fig. 16, 1610, and ¶ 146).
The Examiner finds that in Cote’s context model, “logic for calculating an
updated probability . . . may implement a state machine that transitions
between a finite number of context model states.” Ans. 8 (emphasis
omitted) (citing Cote ¶ 115).
Appellant, in turn, does not rebut the Examiner’s additional reasoning
and findings from Cote articulated in the Answer. Consequently, Appellant
does not persuade us of error in these additional factual findings or in the
Examiner’s conclusion of obviousness.
For the reasons discussed, Appellant does not persuade us of error in
the Examiner’s obviousness rejection of independent claim 1. We,
therefore, sustain the Examiner’s rejection of that claim, as well as the
rejection of dependent claims 2, 4–13, and 15–27, which Appellant does not
argue separately with particularity. See generally Appeal Br. 11–15
(nominally arguing claims 2, 4–13, and 15–27 without providing additional
specific arguments).

Appeal 2019-004665
Application 14/277,172

9
CONCLUSION
We affirm the Examiner’s decision to reject claims 1, 2, 4–13, and
15–27 under 35 U.S.C. § 103.
DECISION SUMMARY
In summary:
Claims
Rejected
35 U.S.C. § Reference(s)/Basis Affirmed Reversed
1, 2, 4–13,
15–27
103 He, Cote 1, 2, 4–13,
15–27

TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).
AFFIRMED