Ex Parte Zhang et al

Patent Trials and Appeals Board

Feb 5, 2019

13533706 - (D) (P.T.A.B. Feb. 5, 2019)

UNITED STA TES p A TENT AND TRADEMARK OFFICE
APPLICATION NO. FILING DATE
13/533,706 06/26/2012
14870 7590 02/07/2019
KNOBBE, MARTENS, OLSON & BEAR, LLP
QUALCOMM
2040 Main Street
Fourteenth Floor
Irvine, CA 92614
UNITED ST A TES OF AMERICA
FIRST NAMED INVENTOR
Xiaopeng Zhang
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.
QCMLF.234A I 113301 7510
EXAMINER
KWAN,MATTHEWK
ART UNIT PAPER NUMBER
2482
NOTIFICATION DATE DELIVERY MODE
02/07/2019 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):
ocpat_uspto@qualcomm.com
efiling@knobbe.com
jayna.cartee@knobbbe.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte XIAOPENG ZHANG, XIAOKANG ZHANG, JILAI WANG,
XIAOMING ZHOU, GANG QIU, LIANG SHEN, and CHEN FENG
Appeal2018-004773
Application 13/533,706
Technology Center 2400
Before JOHN A. EV ANS, JASON J. CHUNG, and JASON M. REPKO,
Administrative Patent Judges.
EV ANS, Administrative Patent Judge.
DECISION ON APPEAL
Appellants 1 seek our review under 35 U.S.C. § 134(a) from the
Examiner's final rejection of Claims 1, 3-17, and 19-36. App. Br. 1. We
have jurisdiction under 35 U.S.C. § 6(b). We REVERSE. 2
1 Appellants state the real party in interest is Qualcomm Incorporated. App.
Br. 3.
2 Rather than reiterate the arguments of the Appellant and the Examiner, we
refer to the Appeal Brief (filed January 16, 2018, "App. Br."), the Reply
Brief (filed April 4, 2018, "Reply Br."), the Examiner's Answer (mailed
February 22, 2018, "Ans."), the Final Action (mailed June 27, 2017, "Final
Act."), and the Specification (filed June 26, 2012, "Spec.") for their
Appeal2018-004773
Application 13/533,706
STATEMENT OF THE CASE
The claims relate to systems and methods for detecting a live human
face in an image. See Abstract.
INVENTION
Claims 1, 16, 29, 35, and 36 are independent. An understanding of
the invention can be derived from a reading of illustrative Claim 1, which is
reproduced below:
1. A method for detecting a live human face in an image,
compnsmg:
receiving multispectral image data that includes a human
face, said multispectral image data comprising visible light
image data and near-infrared (NIR) image data;
processing the multispectral image data to detect the
human face;
associating the detected human face in the visible light
image data with the detected human face in the NIR image data;
and
determining whether the detected human face is a live
human face based at least in part on a first reflectance
difference and a second reflectance difference,
wherein the first reflectance difference is the difference
between a reflectance value in a portion of the detected human
face in the NIR image data and a reflectance value of a first
color in the portion of the detected human face in the visible
light image data, and
wherein the second reflectance difference is the
difference between the reflectance value of the first color and a
respective details.
2
Appeal2018-004773
Application 13/533,706
reflectance value of a second color in the portion of the detected
human face in the visible light image data.
Pavlidis
Pillman
Polonskiy
References and Rejections3
US 2003/0053664 Al Mar. 20, 2003
US 2007 /0248330 Al Oct. 25, 2007
US 2007 /0268485 Al Nov. 22, 2007
Yi D., Face Matching Between Near Infrared and Visible Light Images, Lee
S. W., Li S.Z. (eds.) Advances in Biometrics, ICB 2007, Lecture Notes in
Computer Science, vol. 4642, p. 523-530, Springer, Berlin, Heidelberg
(2007).
Kanazawa Y., Human Skin Detection by Visible and Near-Infrared Imaging,
MVA20Il IAPR Conference on Machine Vision Applications, 503-507,
Nara, JAPAN (2011).
The claims stand rejected as follows:
1. Claims 1, 3, 4, 7, 13-17, 19-22, and 26-35 stand rejected under 35
U.S.C. § I03(a) as being unpatentable over Pavlidis, Polonskiy, and
Yi. Final Act. 2-17.
3 The application was examined under the pre-AIA first to invent provisions.
3
Appeal2018-004773
Application 13/533,706
2. Claims 8-12, and 23-25 stand rejected under 35 U.S.C.
§ 103(a) as being unpatentable over Pavlidis, Polonskiy, Yi, and
Pillman. Final Act. 17-21.
3. Claim 36 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Pavlidis, Polonskiy, Yi, and Kanazawa. Final Act.
21-24.
ANALYSIS
We have reviewed the rejections of Claims 1, 3-17, and 19-36 in light
of Appellants' arguments that the Examiner erred. We consider Appellant's
arguments seriatim, as they are presented in the Appeal Brief, pages 8-21.
CLAIMS 1, 3-17, AND 19-36: OBVIOUSNESS OVERPAVLIDIS, POLONSKIY,
YI, PILLMAN, AND KANAZAWA.
Appellants argue all claims and each ground of rejection as a group.
App. Br. 8. Appellants argue the claims with respect to the limitations of
exemplary Claim 1. See App. Br. 9. Therefore, we decide this appeal on the
basis of Claim 1, and refer to the rejected claims collectively herein as "the
claims." See 37 C.F.R. § 4I.37(c)(l)(iv); In re King, 801 F.2d 1324, 1325
(Fed. Cir. 1986).
Claim 1 recites, inter alia, "receiving multispectral image data that
includes a human face, said multispectral image data comprising visible light
image data and near-infrared (NIR) image data," and "processing the
multispectral image data to detect the human face." Independent Claims 16,
29, 35, and 36 contain commensurate recitations.
4
Appeal2018-004773
Application 13/533,706
The Examiner finds Pavlidis teaches a facial recognition system that
uses a reflectance difference comparison using visible light and infrared
data. Final Act. 3 ( citing Pavlidis ,r,r 20, 62, 66, and 68).
Appellants contend Pavlidis does not teach or suggest that any portion
of the visible data of the image data representative of at least a portion of the
person's face is compared to the near-infrared data of the image data to
detect a person's face. App. Br. 10.
The Examiner finds "Pavlidis discloses detection of artificial face
parts as opposed to a natural (i.e. live) face in [O 117], a reflectance
difference comparison in [0020] and a facial recognition system that uses
visible light and infrared data ([0062], [0066] and [0068])." Ans. 3. The
Examiner further finds "Pavlidis does not explicitly disclose detecting a live
human face or a reflectance difference comparison between visible light and
NIR image data," but Polonskiy teaches "associating the detected human
face in the visible light image data with the detected human face in the NIR
image data ([0062]." Id.
The Examiner finds
[i]t would have been obvious to one of ordinary skill in
the art at the time the invention was made to modify Pavlidis' s
method with the detection of a live human face and comparison
ofNIR and visible light reflectance data as taught by Polonskiy
to be able to detect a person wearing a disguise (Polonskiy
[0008]). Ans. 3--4.
Appellants reply that the claim language requires more than "only [] a
comparison of reflectance differences in 'a portion of the detected human
face."' Reply Br. 4 ( quoting Ans. 25). Appellants argue
5
Appeal2018-004773
Application 13/533,706
Claim 1 requires "determining whether the detected human face
is a live human face based at least in part on a first reflectance
difference and a second reflectance difference, wherein the first
reflectance difference is the difference between a reflectance
value in a portion of the detected human face in the NIR image
data and a reflectance value of a first color in the portion of the
detected human face in the visible light image data." Reply Br.
4--5.
We do not understand the references to teach comparing multispectral
image data comprising visible and NIR data, as found by the Examiner. The
Examiner finds "Pavlidis discloses detection of artificial face parts as
opposed to a natural (i.e. live) face in [0117]." Ans. 3.
Whereas the claims require the detection of a face by comparing
multispectral image data comprising visible and NIR bands, Pavlidis' facial
detection method does not use visible light:
Facial signatures are less variable in the near-infrared which
aids significantly in such detection. Further, illumination in the
scene can be maintained at an optimal level through a feedback
control loop that includes a near-infrared illuminator. In
addition, since near-infrared light is invisible to the human eye,
the face detection can be performed in an unobtrusive and
covert manner. Such advantages in combination with the
unique reflectance characteristics of the human skin in the near-
infrared spectrum allow for simple and reliable algorithmic
based face detection.
Pavlidis, ,r 17 (cited by the Examiner).
The Examiner further cites Pavlidis for teaching "a reflectance
difference comparison in [0020]." Ans. 3. Contrary to the Examiner,
whereas Pavlidis teaches: "a difference in reflectance for human skin in the
first bandwidth relative to the second bandwidth is greater than a difference
in reflectance for objects in the background other than human skin in the
6
Appeal2018-004773
Application 13/533,706
first bandwidth relative to the second bandwidth." Pavlidis ,r 20 ( cited by
the Examiner). However, Pavlidis discloses the multispectral image is
comprised of first and second IR, not visible, bands:
In one embodiment of the method, a first image output is
provided that is representative of reflected light of a scene in a
first bandwidth within a reflected infrared radiation range ... In
addition, a second image output representative of reflected light
of a scene in a second bandwidth within a reflected infrared
radiation range is provided.
Pavlidis, ,r 19.
The Examiner further cites Pavlidis for "a facial recognition system
that uses visible light and infrared data ([0062], [0066] and [0068])." Ans.
3. Pavlidis, paragraph 62, relates to facial detection (not recognition)
systems using thermal IR, not visible light: "[i]n addition, such methods and
systems may be used in combination with other systems and methods such
as those that employ thermal infrared detection." As the Examiner finds,
Pavlidis, paragraph 62 relates to visible light facial recognition. However,
the claims recite, inter alia, "processing the multispectral image data to
detect the human face."
The Examiner finds Pavlidis fails to teach detecting a live, human
face, nor does Pavlidis teach comparing a reflectance difference between
visible and near infrared (NIR) image data. Id. The Examiner cites
Polonskiy for these teachings. Ans. 3 ( citing Polonskiy ,r,r 40, 62, and 63).
Contrary to the Examiner, Polonskiy discloses a visible light
companson:
FIG. 4 shows the difference between normalized reflectance for
living and dead skin. The wavelengths corresponding to the
largest difference between the normalized reflectance of dead
7
Appeal2018-004773
Application 13/533,706
and live skins are the best for liveness detection. From FIG. 4 it
is easy to conclude that these wavelengths are in the blue ( 450-
480 nm) and red (600-700 nm) 4 ranges of the spectrum when
the reference point is selected at 530 nm.
Polonskiy, ,r 40 (cited by the Examiner). Figure 3 of Polonskiy (cited
by the Examiner) "depicts the ratios of reflectivities in all spectral bands to
the reflectivity at the reference wavelength 530 nm. It can be easily seen
that the ratio ( or normalized reflectance Rnorm) is almost constant in the
visible spectral range for dead skin Rnorm = 1 ± 0 .1, while for living skin it
changes in a significantly broader range l