Ex Parte Ganapathi et al

Patent Trial and Appeal Board

Mar 31, 2016

12712871 (P.T.A.B. Mar. 31, 2016)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
121712,871 02/25/2010
40581 7590 04/04/2016
CRAWFORD MAUNU PLLC
1150 NORTHLAND DRIVE, SUITE 100
ST. PAUL, MN 55120
FIRST NAMED INVENTOR
Hariraam V arun Ganapathi
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.
STFD.246PA (S09-319) 3564
EXAMINER
SECK, ABABACAR
ART UNIT PAPER NUMBER
2122
NOTIFICATION DATE DELIVERY MODE
04/04/2016 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):
USPTO-patent@ip-firm.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte HARIRAAM V ARUN GAN AP ATHI,
CHRISTIAN PLAGEMANN, SEBASTIAN THRUN,
and DAPHNE KOLLER
Appeal2014-002804
Application 12/712,871
Technology Center 2100
Before: ELENI MANTIS MERCADER, JEFFREY A. STEPHENS, and
STACY B. MARGOLIES, Administrative Patent Judges.
MANTIS MERCADER, Administrative Patent Judge.
DECISION ON APPEAL
Appeal2014-002804
Application 12/712,871
STATE~vfENT OF CASE
Appellants appeal under 35 U.S.C. § 134(a) from a rejection of claims
1-20. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm.
THE INVENTION
The claimed invention is directed to determining location probabilities
for a plurality of object parts by identifying, from image data obtained from
a depth sensor, features of the object parts. A processing circuit selects a set
of poses for at least one object based upon determined location probabilities
and generates modeled depth sensor data by applying the selected set of
poses to a model of the at least one object. The processing circuit selects a
pose for the at least one object based upon a probabilistic comparison
between the data obtained from the depth sensor and the modeled depth
sensor data. Abstract.
Claim 1, reproduced below, is illustrative of the claimed subject
matter:
1. A system for tracking at least one object articulated in three-
dimensional space using data obtained from a depth sensor, the
system comprising:
at least one processing circuit configured and arranged to:
determine location probabilities for a plurality of object parts by
identifying, from image data obtained from the depth sensor,
features of the object parts;
select a set of poses for the at least one object based upon the
determined location probabilities and based upon identified
feature detections that propose moving a feature to a position
where no other feature currently exists;
generate modeled depth sensor data by applying the selected set
of poses to a model of the at least one object; and
2
Appeal2014-002804
Application 12/712,871
select a pose for the at least one object model-based based upon
a probabilistic comparison between the data obtained from the
depth sensor and the modeled depth sensor data.
REFERENCES
The prior art relied upon by the Examiner in rejecting the claims on
appeal is:
Freeman
Rehg
Jojic
Dantwala
Janssen
Geiss
us 6, 115,052
US 6,269,172 Bl
US 6,674,877 Bl
US 2003/0197712 Al
US 2007 /0299559 Al
US 2010/0197399 Al
Sept. 5, 2000
July 31, 2001
Jan. 6,2004
Oct. 23, 2003
Dec. 27, 2007
Aug. 5, 2010
Devi Parikh et al., Feature-based Part Retrieval for Interactive 3D
Reassembly, Carnegie Mellon University, Intel Research Pittsburgh.
REJECTIONS
The Examiner made the following rejections:
Claims 1-5 stand rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Geiss.
Claim 6 stands rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Geiss and further in view ofDantwala.
Claim 7 stands rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Geiss and Rehg.
Claim 8 and 9 stands rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Geiss and Janssen.
Claims 10-14 and 18 stand rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Parikh.
3
Appeal2014-002804
Application 12/712,871
Claim 15 stands rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Parikh and Dantwala.
Claim 16 stands rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Parikh and Rehg.
Claim 17 stands rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Parikh and Janssen.
Claims 19 and 20 stand rejected under 35 U.S.C § 103(a) as being
unpatentable over Jojic in view of Parikh and Freeman.
ISSUES
The pivotal issues are whether the Examiner erred in finding that the
combination of Jojic in view of Geiss teaches the limitation of "select a
pose" as recited in claim 1 and the combination of Jojic in view of Parikh
teaches the limitation of "selecting a pose" as recited in claim 10.
ANALYSIS
Appellants argue that the assignment of a pixel to a body part, as is
taught by Jojic, is not the same as selection of a pose because there is no
selection of a "posture," but rather a determination of a location of an object
(i.e., body part) without assessing the set of poses to depth sensor data (App.
Br. 4).
We do not agree with Appellants' argument. We agree with the
Examiner's finding that Jojic teaches a selection of a posture by teaching
that the "Gaussian model then uses the 3-D points for each pixel to compute
a likelihood estimate of a posture of the articulated structure" (emphasis
added) (Jojic col. 7, 11. 3--4; see also Ans. 23). Jojic teaches that the 3-D
4
Appeal2014-002804
Application 12/712,871
Gaussian model can be used as a rough model of a body part and a depth
processor 214 computes range image data (see Ans. 23; Jojic col. 6, 1. 23-
col. 7, 1. 27). We further agree with the Examiner that Geiss teaches
selecting a pose initially and then selecting a more accurate pose based on
the analysis of a depth image (see Final Act. 3, Geiss paras. 59 and 78).
Thus, both Jojic and Geiss teach a selection of a pose.
Appellants further argue that the coarse level of analysis, which
provides a rough or loose textured mapping as taught by Geiss, would not
provide dense disparity maps that are required by Jojic (App. Br. 6-7).
Appellants further argue that the references do not result in an enabled
combination, the modification amounts to an "obvious to try" allegation, and
there is no motivation to combine Jojic with Parikh in rejecting claim 10
(App. Br. 8-10).
We are not persuaded by Appellants' arguments. Appellants'
arguments are unpersuasive because it is well settled that "a determination of
obviousness based on teachings from multiple references does not require an
actual, physical substitution of elements." In re Mouttet, 686 F.3d 1322,
1332 (Fed. Cir. 2012) (citations omitted). Nor is the test for obviousness
whether a secondary reference's features can be bodily incorporated into the
structure of the primary reference. See In re Keller, 642 F.2d 413, 425 (Fed.
Cir. 1981). We agree with the Examiner's findings in the Final Office
Action that Jojic teaches a pose being selected based upon the probabilistic
comparison of data obtained from the depth sensor and the modeled depth
sensor (Final Act. 3; see Jojic Fig. 3A; col. 6, 1. 51---col. 7, 1. 15, col. 12, 1. 7-
col. 13, 1. 15). We further agree with the Examiner that Geiss teaches
selecting a more accurate pose after an initial pose is selected (Final Act. 3;
5
Appeal2014-002804
Application 12/712,871
Geiss paras. 59, 78). The two algorithms do not have to be physically
substituted or combined, but rather one skilled in the art would know how to
further improve the teaching of a selected pose by Jojic to a more accurate
pose as taught by Geiss, and thereby propose moving a feature to a position
where no other feature currently exists (i.e., moving the more inaccurate
pose selection to a more accurate pose). We also agree with the Examiner's
motivation articulated in the Final, regarding claim 10, wherein the
modification would provide compatibility that enables the construction of an
interactive system for 3D reassembly, where the user can easily assemble a
desired object from a large collection of pieces (many of which are
irrelevant) by iteratively selecting compatible parts (see Final Act. 11-12).
An artisan is presumed to possess both skill and common sense. See
KSR Int'! Co. v. Teleflex Inc., 550 U.S. 398, 421 (2007) ("A person of
ordinary skill is also a person of ordinary creativity, not an automaton.").
Furthermore, Appellants have not presented any evidence demonstrating that
the modification would have been "uniquely challenging or difficult for one
of ordinary skill in the art." See Leapfrog Enters., Inc. v. Fisher-Price, Inc.,
485 F.3d 1157, 1162 (Fed. Cir. 2007) (citing KSR, 550 U.S. at 418).
Accordingly, we affirm the Examiner's rejections of claims 1 and 10,
and, for the same reasons, the rejections of claims 2-9 and 11-20 not argued
separately.
CONCLUSION
The Examiner did not err in finding that the combination of Jojic in
view of Geiss teaches the limitation of "select a pose" as recited in claim 1
6
Appeal2014-002804
Application 12/712,871
and the combination of Jojic in view of Parikh teaches the limitation of
"selecting a pose" as recited in claim 10.
The Examiner did not err in combining Jojic with Geiss in rejecting
claim 1 or Jojic with Parikh in rejecting claim 10.
DECISION
For the above reasons, the Examiner's rejections of claims 1-20 are
affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(l )(iv).
AFFIRMED
7