Ex Parte Prakash et al

Patent Trial and Appeal Board

Sep 6, 2016

11452087 (P.T.A.B. Sep. 6, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR
111452,087 06/12/2006 Adityo Prakash
15610 7590 09/06/2016
Okamoto & Benedicto LLP
PO Box 641330
San Jose, CA 95164-1330
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www .uspto.gov
ATTORNEY DOCKET NO. CONFIRMATION NO.
10006.002091
(Al424CIPC1)
ART UNIT
2486
MAILDATE
09/06/2016
1573
EXAMINER
VO,TUNGT
PAPER NUMBER
DELIVERY MODE
PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte ADITYO PRAKASH and ENIKO FODOR
Appeal2015-003988
Application 11/452,087
Technology Center 2400
Before CARL W. WHITEHEAD JR., KARA L. SZPONDOWSKI, and
MICHAEL J. ENGLE, Administrative Patent Judges.
SZPONDOWSKI, Administrative Patent Judge.
DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from the Examiner's
Final Rejection of claims 1-8, 10-13, 16-18, and 20, the only claims
pending in the application. We have jurisdiction under 35 U.S.C. § 6(b ).
We REVERSE.
Appeal2015-003988
Application 11/452,087
STATEMENT OF THE CASE
Appellants' invention is directed to a segment-based encoding system
including segment-specific metadata. (Spec. i-f 2.) Claim 1, reproduced
below, is illustrative of the claimed subject matter:
1. An encoding apparatus for encoding uncompressed video data
to form compressed video data which is decodable to form at least an
approximation of the uncompressed video data, wherein the video
data comprises a plurality of image frames comprising key frames and
nonkey frames, wherein a nonkey frame is a frame that is encoded
with reference to segmentation of other frames (reference frames), and
wherein a segmentation of a reference frame is an assignment of some
or all of the pixels of the reference frame to segments based on at least
one of pixel color values of the pixels and location of the pixels in the
reference frame, the apparatus comprising:
a) a segmenter that generates a segmentation of a reference
frame;
b) a motion matcher that matches a segment of the reference
frame to an area of a current frame being encoded when the
current frame is a nonkey frame;
c) a metadata associator that creates a metadata element
associated with the segment of the reference frame and
further associated with the matched area of the current
frame, wherein the metadata element is inherited, such that
when a segment associated with the matched area in the
current frame is broken up into a group of segments in a
subsequent frame, the segments in the group all inherit the
metadata element; and
d) compressed video data outputter that outputs compressed
video data including at least compressed key frames, the
kinetic information and the metadata element.
2
Appeal2015-003988
Application 11/452,087
REJECTIONS
Claims 1--4, 6-8, 10, and 12-14 stand rejected under 35 U.S.C.
§ 102(e) as being anticipated by Mead (US 6,683,993 Bl; issued Jan. 27,
2004).
Claims 5 and 16 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Mead and Chen (US 6,278,466 B 1;
issued Aug. 21, 2001 ).
Claim 18 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Mead and Citta (US 5,285,276; issued
Feb. 8, 1994).
Claim 20 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over the combination of Mead, Citta, and Chen.
Claims 11and17 stand rejected under 35 U.S.C. § 102(e) as being
anticipated by Jain et al. (US 6,567,980B1; issued May 20, 2003).
ANALYSIS
Claim 1
Issue 1: Did the Examiner err in finding Mead discloses "the
metadata element is inherited, such that when a segment associated with the
matched area in the current frame is broken up into a group of segments in a
subsequent frame, the segments in the group all inherit the metadata
element," as recited in independent claim 1?
The Examiner relies on elements 44, 46, 58, 60, and 62 of Figure 2,
elements 58, 60, and 62 of Figure 1, and column 1, lines 45-57 and column
3, lines 50-55 of Mead to disclose the disputed limitation. (Final Act. 3--4;
Ans. 13, 15-16.) Appellants argue Mead does not disclose that symbolic
3
Appeal2015-003988
Application 11/452,087
code 58 is inherited. (App. Br. 9-10; Reply Br. 2.) Specifically, Appellants
contend Mead discloses that "multiple features of an object are used to
produce a symbolic code corresponding to the closest recognized object."
(App. Br. 10.) Further, Appellants argue that although "Mead recites that
objects (including a human face, a hand, an arm, and a background) may be
extracted by a feature extractor ... [ f]eature extraction and metadata
inheritance are entirely different to one of ordinary skill in the art at the time
of filing." (App. Br. 10.)
We are persuaded by Appellants' arguments. Mead describes an
object encoder 44 that includes an image segmenter 46, which subdivides an
image into at least one object. (Mead col. 3, 11. 45--47.) An object can be
representative of a constituent part of the image, a region of the image, or
another entity of interest in the image. (Mead col. 3, 11. 48-51.) The object
encoder in Mead extracts the object. (Mead col. 3, 11. 43--44.) The object
encoder further includes a feature extractor 48. (Mead col. 3, 11. 55-56.) A
classifier 50 compares the extracted object to a set of generic objects located
in a generic library 52 for object recognition. (Mead col. 3, 1. 64- col. 4, 1.
2.) The classifier then produces a symbolic code 58 corresponding to the
closest recognized object in the generic library based upon the at least one
feature quantity. (Mead col. 4, 11. 2-15.)
While we agree with the Examiner's general finding that the symbolic
code 58 of Mead discloses the claimed metadata element, (Final Act. 3, Ans.
15), we agree with Appellants that the Examiner has not established that
Mead discloses "wherein the metadata element is inherited, such that when a
segment associated with the matched area in the current frame is broken up
into a group of segments in a subsequent frame, the segments in the group
4
Appeal2015-003988
Application 11/452,087
all inherit the metadata element." The Examiner's rejection lacks sutlicient
findings as to how the symbolic code is inherited, as recited in the claim.
For example, the Examiner has not shown that after a symbolic code for an
object has been created, the object is subsequently "broken up" into smaller
segments and that each segment "inherits" the same symbolic code.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 102(e)
rejection of independent claim 1. For the same reasons, we do not sustain
the Examiner's 35 § 102(e) rejection of dependent claims 2-8 and 10.
Claims 12 and 18
Issue 2: Did the Examiner err in finding Mead discloses "wherein the
metadata element specifies a response to an input indicating a user selection
of the area of the current frame associated with the metadata element," as
recited in independent claim 12? Did the Examiner err in finding Mead
teaches or suggests a similar recitation in independent claim 18?
The Examiner relies on the Abstract of Mead, particularly the
"segment selector for identifying a signal segment from a group of speech,
audio, video and graphic signals," to disclose the "response to an input
indicating a user selection of the area." (Final Act. 6; Ans. 16-17.) The
Examiner further finds "[t]he disclosure clearly shows the segment selector
is treated as a user selection and they have the same function" (Ans. 13) and
"the decoder can send a request as an input user selection to the encoder to
encode[] the particular area of the current image or picture." (Ans. 17.)
Appellants argue "[i]dentifying a signal type (speech, audio, video or
graphic) per Mead is entirely different from 'an input indicating a user
selection of the area."' (App. Br. 11; see also Reply Br. 3--4.) Appellants
5
Appeal2015-003988
Application 11/452,087
further contend "[r]egarding the decoder sending a request as an input user
selection to the encoder, there is no such teaching in Mead." (Reply Br. 4.)
We agree with Appellants that the Examiner has not established Mead
discloses "the metadata element specifies a response to an input indicating a
user selection of the area of the current frame associated with the metadata
element." The Examiner's rejection lacks sufficient findings as to how the
segment selector discloses the response to an input indicating a user
selection of the area. Moreover, the Examiner has provided no support or
citation for the findings "[t]he disclosure clearly shows the segment selector
is treated as a user selection and they have the same function" and "the
decoder can send a request as an input user selection to the encoder to
encode[] the particular area of the current image or picture." For example,
the Examiner has not identified who or what the "user" is, let alone cited
evidence supporting such a finding.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 102(e)
rejection of independent claim 12 and 35 U.S.C. § 103(a) rejection of
independent claim 18. For the same reasons, we do not sustain the
Examiner's 35 U.S.C. § 102(e) rejection of dependent claims 13, 14, and 20,
and the Examiner's 35 U.S.C. § 103(a) rejection of dependent claim 16.
Claim 11
Issue 3: Did the Examiner err in finding Jain discloses "a segmenter
that ... further generates a segmentation hint which is used by a decoder so
as to allow the decoder to reconstruct a segmentation of an original version
of the frame rather than a segmentation of a reconstructed version of the
frame," as recited in independent claim 11?
6
Appeal2015-003988
Application 11/452,087
The Examiner relies on elements 470, 472, 474, 476, and 478 of
Figure 8 and element 914 in Figure 16 to disclose the disputed limitation.
(Final Act. 8; Ans. 19.) Specifically, the Examiner relies on elements 470-
478 of Figure 8 to disclose the "segmentation hint" and element 914 in
Figure 16 to disclose "as to allow the decoder to reconstruct a segmentation
of an original version of the frame rather than a segmentation of a
reconstructed version of the frame." (Final Act. 8; Ans. 19.)
Appellants argue the portions of Jain relied upon by the Examiner do
not disclose the disputed limitation, particularly the "segmentation hint."
(App. Br. 13-14.) Appellants further argue "Jain does not even distinguish
between an original version of a frame and a reconstructed version of a
frame." (Reply Br. 5.)
We are persuaded by Appellants' arguments that the Examiner has not
established Jain discloses the segmenter "further generates a segmentation
hint which is used by a decoder so as to allow the decoder to reconstruct a
segmentation of an original version of the frame rather than a segmentation
of a reconstructed version of the frame." The elements in Figure 8 relied
upon by the Examiner to disclose the "segmentation hint" are a Start-up
Extensibility Initialization manager (470), Feature Extractor registration
(471), Event registration (474), Metadata track registration (476), and
Output Filter registration (478). (Jain Fig. 8, col. 8, 11. 24--47.) The
Examiner's rejection lacks sufficient findings as to how these disclose a
segmentation hint.
Element 914 in Figure 16 describes "output main HTML page
incorporating frames and video play-out helper app." (Jain Fig. 16.) The
Examiner's rejection lacks sufficient findings as to how this discloses
7
Appeal2015-003988
Application 11/452,087
allowing the decoder to reconstruct a segmentation of an original version of
the frame rather than a segmentation of a reconstructed version of the frame.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 102(e)
rejection of independent claim 11.
Claim 17
Issue 4: Did the Examiner err in finding Jain discloses "the metadata
element is inherited, such that when a segment associated with the matched
area in the current frame is broken up into a group of segments in a
subsequent frame, the segments in the group all inherit the metadata
element," as recited in independent claim 17?
Appellants contend Jain does not that disclose segments in a group
inherit the metadata element. (App. Br. 15; Reply Br. 5.) The Examiner
relies on elements 110 and 112 of Figure 1, elements 402-412 of Figure 7,
and elements 470-478 of Figure 8, and Figure 6 to disclose the disputed
limitation. (Final Act. 8; Ans. 20-21.) The Examiner relies on the metadata
tracks (keyframes, cc-text, audio classes, and speech) to disclose the
metadata element, (Ans. 20-21 ), and further finds a "keyframe is a segment
that includes a group of segments in a subsequent frame." (Final Act. 8.)
We agree with Appellants. Although we agree with the Examiner's
general finding that the metadata tracks of Jain disclose the claimed
metadata element, we agree with Appellants that the Examiner has not
established that Jain discloses "wherein the metadata element is inherited,
such that when a segment associated with the matched area in the current
frame is broken up into a group of segments in a subsequent frame, the
segments in the group all inherit the metadata element." The Examiner's
8
Appeal2015-003988
Application 11/452,087
rejection lacks sufficient findings as to how the metadata element is
inherited.
Accordingly, we do not sustain the Examiner's 35 U.S.C. § 102(e)
rejection of independent claim 1 7.
DECISION
For the above reasons, the Examiner's rejection of claims 1-8, 10-13,
16-18, and 20 is reversed. 1
REVERSED
1 In the event of further prosecution, we leave it to the Examiner to evaluate
claim 1 for compliance with 35 U.S.C. § 112, second paragraph, based on
whether there is sufficient antecedent basis for "the kinetic information."
9