Ex Parte 7542610 et al

Board of Patent Appeals and Interferences

Jan 19, 2012

95001307 (B.P.A.I. Jan. 19, 2012)

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.
95/001,307 03/17/2010 7542610 OJOS.P106REX 4791
100109 7590 01/19/2012
Mahamedi Paradice Kreisman LLP (Google, Inc)
550 South Winchester Boulevard
Suite 605
San Jose, CA 95128
EXAMINER
LAROSE, COLIN M
ART UNIT PAPER NUMBER
3992
MAIL DATE DELIVERY MODE
01/19/2012 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 BOARD OF PATENT APPEALS
AND INTERFERENCES
____________

SUPERFISH, INC.
Third Party Requester, Appellant

v.

LIKE.COM, INC.
Patent Owner, Respondent
________

Appeal 2011-006145
Inter partes Reexamination Control 95/001,307
United States Patent 7,542,610 B2
Technology Center 3900
____________

Before SALLY C. MEDLEY, KARL D. EASTHOM, and
KEVIN F. TURNER, Administrative Patent Judges.

TURNER, Administrative Patent Judge.

DECISION ON APPEAL

Third Party Requester Superfish, Inc. (hereinafter “Appellant”)
appeals1 under 35 U.S.C. §§ 134(b) and 315(b)(1) the Examiner’s decision

1 See Third Party Requester’s Appeal Brief filed September 15, 2010,
hereinafter “App. Br.,” at 2; see also Third Party Requester’s Rebuttal Brief
filed January 27, 2011, hereinafter “Rebut. Br.”
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

2
to not adopt Appellant’s proposed rejections and confirm claims 1-252.
Patent Owner Like.com, Inc., (hereinafter “Respondent”) is a party to the
appeal under 35 U.S.C. § 315(a)(2) and disputes the Appellant’s
contentions.3
We have jurisdiction under 35 U.S.C. §§ 6, 134 and 315. We
AFFIRM.

STATEMENT OF THE CASE
This proceeding arose from a corrected request for inter partes
reexamination filed by Steven J. Frank on behalf of Requester, on March 17,
2010, of United States Patent 7,542,610 B2 (“the '610 Patent”), issued to
Salih Burak Gokturk, Munjal Shah, and Azhar Khan on June 2, 2009, based
on United States Application 11/543,758, filed October 3, 2006. That
application claimed benefit to the following applications:
Continuation-In-Part of 11/246,434 Oct. 7, 2005
Continuation-In-Part of 11/246,589 Oct. 7, 2005
Continuation-In-Part of 11/246,741 Oct. 7, 2005
Continuation-In-Part of 11/246,742 Oct. 7, 2005
Provisional 60/679,591 May 9, 2005
Provisional 60/723,349 Oct. 3, 2005
Provisional 60/723,356 Oct. 3, 2005
The '610 patent is the subject of pending litigation styled as Like.com
v. Superfish, Inc., Civ. A. No. 4:09-cv-5805-SBA (N.D. Cal.). (See App. Br.
48). The action was stayed pending the outcome of this reexamination (id.).

2 See Right of Appeal Notice, mailed June 16, 2010, hereinafter “RAN;” see
also Examiner’s Answer, mailed December 29, 2010, hereinafter “Ans.”
3 See Patent Owner’s Respondent Brief filed October 15, 2010, hereinafter
“Resp. Br.”
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

3
We heard oral arguments from the representatives of both the
Appellant and the Respondent, on June 15, 2011, a transcript4 of which is
part of the record.

THE INVENTION
Patentee’s invention relates to digital image processing, and the use of
captured images and indexing of information about those images to allow for
search and selection. (Spec. Col. 1, ll. 46-49 and Abs.)
Exemplary claim 1 on appeal reads as follows:
1. A method for enabling selection of image content items, the
method comprising: using a processor to perform steps comprising:
performing image analysis on a collection of image content
items to obtain information about each image content item;
wherein performing image analysis includes (i) identifying one
or more objects in individual image content items, (ii) determining a
category of each identified object, the category of each object being
one of a plurality of possible object categories; (iii) identifying a set of
features that are specific to the determined category of the identified
object in each image content item, and (iv) determining information
based on the set of features for the determined category that
characterizes the identified object, the information characterizing the
object to be separately identifiable from at least some other identified
objects in the determined category;
storing the information obtained from the image analysis in one
or more data stores;
identifying one or more criteria determined from a text or image
content provided in connection with either a user or programmatically
identified input; and
performing a search operation to identify one or more image
content items that satisfy the one or more criteria of the input using
the stored information in the one or more data stores;

4 Record of Oral Hearing (hereinafter Oral Hr’g Tr.)
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

4
wherein at least some of the collection of image content items
correspond to images of merchandise objects;
wherein performing the search operation includes selecting one
or more image content items of merchandise objects for display with a
document in response to said input specifying a merchandise object in
the document;
wherein selecting one or more images of merchandise objects
for display with the document includes selecting the one or more
images based on a determination that the one or more images of
merchandise objects are similar to the specified merchandise object;
providing a link with each of the selected one or more images
of the merchandise objects, wherein the link is selectable to enable a
user to purchase the merchandise objects from a network site operated
by the merchant.
(App. Br. 40-41, Claims Appx.).

POTENTIAL PRIOR ART REJECTIONS
The prior art references relied upon by Appellant in the proposed
rejections, upon which the non-adoption is appealed, are:
Ritter 2002/0187774 A1 Dec. 12, 2002
Gokturk ‘906 2003/0169906 A1 Sep. 11, 2003
Shin 2003/0195901 A1 Oct. 16, 2003
Ozaki 2005/0078885 A1 Apr. 14, 2005

Higashide JP2002-328925 Nov. 15, 2002
(English translation dated Jan. 25, 2010)

Aryya Gangopadhyay, An Image-Based System for Electronic Retailing, 32
Decision Support Systems 107-116 (2001) (hereinafter “Gangopadhyay”).
James Z. Wang et al., SIMPLIcity: Semantics-Sensitive Integrated Matching
for Picture Libraries, 23(9) IEEE Trans. Pattern Analysis and Machine
Intelligence 947-963 (Sept. 2001) (hereinafter “Wang”).
Yi Li and Linda G. Shapiro, Object Recognition for Content-Based Image
Retrieval, 1-19 (Jan. 2002), available at
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

5
http://www.cs.washington.edu/homes/ shapiro/cbir.html (last visited Oct. 31,
2011) (hereinafter “Li”).
James Liu and Jane You, Smart Shopper: An Agent-Based Web-Mining
Approach to Internet Shopping, 11(2) IEEE Trans. Fuzzy Syst. 226-237
(Apr. 2003) (hereinafter “Liu”).
Appellant proposed rejections of the claims on the following bases,
upon which the non-adoption is appealed:5
A. claims 1, 2, 4-7, and 11-25 under 35 U.S.C. § 103(a) as being
unpatentable over Liu and Gokturk ‘906 (RAN 2-5);
B. claims 1-4, 6-11, 13, 15, 17-19, 21, and 22 under 35 U.S.C.
§ 103(a) as being unpatentable over Gangopadhyay, Li, and Liu
(RAN 11-13);
C. claims 1, 2, 5-11, and 15-25 under 35 U.S.C. § 103(a) as being
unpatentable over Ritter, Li, and Liu (RAN 14-15); and
D. claims 1-25 under 35 U.S.C. § 103(a) as being unpatentable over
Wang, Higashide, and Liu (RAN 6-7).

ISSUES
Appellant argues that the Examiner erred in not adopting the above-
discussed rejections (App. Br. 14-15). Appellant argues that the Examiner
incorrectly identified Liu and Gangopadhyay as not teaching or suggesting
category-specific features within a specific category (App. Br. 14).
Appellant also argues that Gokturk ‘906 teaches facial classification and

5 Appellant also requested consideration of rejections based on Shin and
Ozaki (i.e., Issues “E” and “F”) (App. Br. 13-14), where those rejections
were not properly considered by the Examiner, and thus not before us on
appeal. See Decision on Petition dated March 21, 2011.
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

6
recognition, i.e., category-specific features (App. Br. 16-20), and that Li
expressly discloses analyzing certain useful features depending on the
category of the object (App. Br. 22-24). Appellant makes similar arguments
with respect to Ritter (App. Br. 26-27), and argues that the Examiner erred
in finding that Wang determines the semantic class for the entire image
rather than determining the category of individual objects in that image
(App. Br. 28-31).
Respondent agrees with the Examiner’s decision (Resp. Br. 5), and
disagrees with Appellant’s contentions. Respondent argues that aspects
recited in Gokturk ‘906 are not specific to the determined category of the
identified object, as required by the claims (Resp. Br. 8-9), and that
Gangopadhyay employs color, texture, and shaped features for objects in
every category (Resp. Br. 9-10). Respondent also argues that irrespective of
the teachings of Higashide, they do not cure the noted deficiencies of Wang
(Resp. Br. 12). Appellant disputes the points raised by Respondent (Rebut.
Br. 2-9).
In not adopting the proposed rejections, the Examiner found that Liu
failed to cure the deficiencies of Gokturk ‘906 by teaching the classification
of an object into a given category and then identifying a category-specific
set of features for the object (RAN 4), and that Wang and Higashide fail to
disclose the categorization of different identified objects within an image
(RAN 6-7). The Examiner also did not adopt additional rejections because
Li fails to cure deficiencies of Gangopadhyay by teaching that categorized
objects can be distinguished from other objects within the same category
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

7
(RAN 12), and that Ritter and Li fail to teach or suggest all of the elements
of the independent claims for the same reasons (RAN 14-15).
We have considered in this decision only those arguments that
Appellant and Respondent actually raised in the Briefs. Arguments which
Appellant or Respondent could have made but chose not to make in the
Briefs are deemed to be waived. See 37 C.F.R. § 41.67(c)(1)(vii).
The issues arising from the respective positions of Appellant,
Respondent, and the Examiner, which we consider herein, are:
A. whether the Examiner erred in not adopting the rejection of claims
1, 2, 4-7, and 11-25 under 35 U.S.C. § 103(a) as being
unpatentable over Liu and Gokturk ‘906?
B. whether the Examiner erred in not adopting the rejection of claims
1-4, 6-11, 13, 15, 17-19, 21, and 22 under 35 U.S.C. § 103(a) as
being unpatentable over Gangopadhyay, Li, and Liu?
C. whether the Examiner erred in not adopting the rejection of claims
1, 2, 5-11, and 15-25 under 35 U.S.C. § 103(a) as being
unpatentable over Ritter, Li, and Liu? and
D. whether the Examiner erred in not adopting the rejection of claims
1-25 under 35 U.S.C. § 103(a) as being unpatentable over Wang,
Higashide, and Liu?

Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

8
FINDINGS OF FACT
The record supports the following findings of fact (FF) by a
preponderance of the evidence.
Liu
FF1. Liu is directed to an agent-based Internet shopping system
(Abs.). The system allows for shoppers to select products from
specific categories, such as shirts, shoes, trousers, skirts, etc., and then
compares the selections made with the recommendations list from the
Smart Shopper, based on the customer’s descriptions, including color,
size, style and fitness (p. 235).

Gokturk ‘906
FF2. Gokturk ‘906 is directed to a method and apparatus for
recognizing objects, through the use of depths determined for surface
regions of objects (Abs.). A portion of the process allows for
detection and classification of an object in an image, to determine if
the object is a person (¶¶ [0052]-[0053], [0146]-[0152]; Fig. 10). If
the object is a person, a face is located and facial features are
extracted to identify the person (id.). The system allows for
classification of vehicle occupants as “an adult, a child, a child seat,
an animal, an object, or an empty seat” (¶ [0153]).

Gangopadhyay
FF3. Gangopadhyay is directed to an image-based system for
electronic retailing (Abs.). As part of the retailing process, customers
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

9
are presented with “matching” apparel, which are categorized and
displayed in groups, such as ties, shirts, jackets, trousers, etc. (p. 114,
§5; Fig. 6). The matching apparel is disclosed to have common
attributes among the categories, such as price, textual description,
shape, color, and texture (id.).

Li
FF4. Li is directed to processes of object recognition for
content-based image retrieval (Abs.). Li focuses on specific
embodiments of object class recognition, including boats, vehicles and
buildings (pp. 4-14). The particular embodiment of recognizing a
boat provides a process by which sky and water regions are detected,
and specific colors are also detected to determine the potential
presence of a boat in the image (pp. 4-5, § 3.1; Fig. 1).

Wang
FF5. Wang is directed to an image retrieval and classification
system (Abs.). Wang employs an automatic classification method to
categorize a general-purpose picture library into semantic classes (p.
949, § 1.3.1). The entire image is classified as a graph, photograph,
textured, nontextured, etc., to aid in the searching of picture libraries
(id.).

Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

10
Higashide
FF6. Higashide is directed to an information processing device
that retrieves web site addresses of products which correspond to the
image of the product (Abs.). The system uses a database server that
carries out image interpretation and retrieves multiple candidates for
products inferred by the images (¶ [0024]).

PRINCIPLES OF LAW
“To render a later invention unpatentable for obviousness, the prior art
must enable a person of ordinary skill in the field to make and use the later
invention.” In re Kumar, 418 F.3d 1361, 1369 (Fed. Cir. 2005) (citing
Beckman Instruments, Inc., v. LKB Produkter AB, 892 F.2d 1547, 1551 (Fed.
Cir. 1989) and In re Payne, 606 F.2d 303, 314-15 (CCPA 1979)).
“Section 103 forbids issuance of a patent when ‘the differences
between the subject matter sought to be patented 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).
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

11
ANALYSIS
ISSUE A
Liu and Gokturk ‘906, Independent Claims 1 and 15
Appellant argues that Liu discloses the sub-steps of claim 1 that
involve the identity of category-specific features because Liu discloses color
and style, the same features recited in the '610 Patent (App. Br. 16).
Respondent argues that the Examiner did not acknowledge the specific
teachings of Liu because Liu does not disclose that limitation (Resp. Br. 6)
and instead provides the identification of a single object with multiple
perspective images (id.). Respondent also argues that while Liu discusses
color and style, it does not provide that they are category-specific features
(Resp. Br. 7). We agree with Respondent and the reasoning adopted by the
Examiner.
While Liu discusses color and style (FF1), Liu does not disclose such
aspects as category-specific features. Claim 1, for example, recites
“identifying a set of features that are specific to the determined category of
the identified object in each image content item.” Appellant appears to
accept that a general feature, applied among many categories is not
necessarily a category-specific feature (App. Br. 14-15; see also Oral Hr’g
Tr. 6-7)6. Liu applies the features of color and style to all objects, such that
they would not be category-specific and would not meet the limitations of
the claims. Thus, we agree with the Examiner that even if the features in Liu

6 “MR. YONAY: . . . So when I read specific in the claim, I read it to mean
that it refers specifically to – this feature refers to this category and perhaps
not to other ones, but it doesn’t necessarily mean unique” (Oral Hr’g Tr. 6).
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

12
and the '610 Patent are disclosed in both, they are not applied in the same
way to meet the claim limitations.
Appellant also argues that Gokturk ‘906 applies the specific sub-steps
in facial classification and recognition (App. Br. 17-19). Respondent argues
that the methodology in Gokturk ‘906 merely locates a person’s face, if
applicable, and performs image recognition, but that is not equivalent to
identifying a set of features that are specific to the determined category of
the identified object (Resp. Br. 8). We generally agree with Respondent.
While Appellant argues that Gokturk ‘906 classifies objects, including
adult human being, child, infant, pet or non-animate object (App. Br. 18),
further identification is only performed once a face is located (FF2). If the
object classified is a pet or an inanimate object, no further identification is
disclosed, and if the object is classified as human, only facial recognition is
performed, i.e., the same set of features for all characterized humans. Thus,
for certain categories of image content items, no additional set of features
would be identified. Claim 1, for example, requires “identifying a set of
features that are specific to the determined category of the identified object”
(emphasis added). Thus, in the case of Gokturk ‘906, no features would be
identified, save for a single class of objects, i.e. for a single category. We
find that claims 1 and 15 require multiple categories, i.e., “plurality of
possible object categories,” with each category having category-specific
features. We do not find Gokturk ‘906, in combination with Liu, teaches or
suggests identifying category-specific features for each identified and
characterized object.
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

13
Implicit in this finding, we also take into account that claim 1, as well
as claim 15, does not require the identification of multiple objects, but only
requires “(i) identifying one or more objects” (Rebut. Br. 6). Thus, one
could argue that if only one object is identified, and it is a human, then the
rest of the claim is satisfied. However, the claims also require multiple
categories as found above. Gokturk ‘906 does disclose multiple categories
(pet or non-animate object), where the non-human categories have no
category-specific features. Thus, Gokturk ‘906 provides a single category
with features and other categories with no features upon which discernment
from other objects in the same category can be made. As discussed above,
we find claims 1 and 15 as requiring multiple categories, each with category-
specific features. We do not find that one of ordinary skill in the art would
have been lead to adopt an image analysis system with multiple categories,
each with category specific features, in view of Liu and Gokturk ‘906.
Appellant also appears to argue that one of ordinary skill in the art
would have performed equivalent feature identification for non-persons, as
disclosed in Gokturk ‘906 (Fig. 10, step 1030, “Perform Other Action”) to
meet the requirements of claims 1 and 15 (App. Br. 16-18). However,
Appellant has not articulated a clear rationale for doing this which is
supported by Gokturk ‘906, Liu, or elsewhere. While other actions may be
dictated by the disclosure of Gokturk ‘906 upon the characterization of a
non-human object, we do not find the functionality of identifying category-
specific features nor separately distinguishing from other objects in the same
category based on those features, as required by claims 1 and 15.
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

14
Appellant also identifies Gokturk ‘906 as disclosing category-specific
features of “age group” which is the same as features disclosed in the '610
Patent (App. Br. 19-20). Respondent argues Gokturk ‘906 does not teach
age group as a feature and at most utilizes it as a category (Resp. Br. 9). We
agree with Respondent.
We do not find it dispositive that the '610 Patent discloses the use of
age groups as a category-specific feature because the disclosure of that
Patent cannot be employed to find its claims to be unpatentable. The central
question is whether one of ordinary skill in the art would have found the use
of age group to be a category-specific feature in view of Gokturk ‘906 and
Liu, not in view of the '610 Patent. We find no credible motivation for such
a view from the references cited, nor the rejection as proffered. We further
agree with Respondent that Gokturk ‘906 uses age group as a means for
characterization, i.e., determining whether adult, child, or infant, and not as a
feature within a category. We do not find Gokturk ‘906 as teaching that an
object is characterized as human and then further characterized as an adult or
not; rather, persons are identified, and facial recognition may be performed,
but no further features are determined.
Thus, with respect to the Examiner’s decision to not adopt the
rejection of claims 1 and 15 over Liu and Gokturk ‘906, we do not find that
the Examiner erred in that non-adoption.

Dependent claims 2, 4-7, 11-14, and 16-25
We acknowledge that Appellant specifically argues that dependent
claims 2, 4-7, 11-14, and 16-25 should be rejected over the combination of
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

15
Liu and Gokturk ‘906 (App. Br. 20-22), but concur with the Examiner’s
non-adoption of their rejection for the same reasons provided supra with
respect to the non-adoption of the rejection of independent claims 1 and 15.

ISSUE B
Gangopadhyay, Li, and Liu, Independent Claims 1 and 15
Appellant argues that Gangopadhyay describes characterizing objects
by visual features into groups which are “nearly identical to the examples
provided in the '610 Patent” (App. Br. 22). Respondent argues that the
Examiner was correct in finding that Gangopadhyay employs color, texture,
and shape features for objects in every category (Resp. Br. 10). We agree
with Respondent.
Gangopadhyay groups objects into categories, such as shirts, trousers,
etc., and further distinguishes them by color (FF3). Even if these are the
same or similar categories recited in the '610 Patent, there must also be
teachings or suggestions of features specific to the determined category.
However, nothing in Gangopadhyay indicates that color is category specific.
If, for example, only certain colors were available in trousers and other color
only available for shirts, then the color features could be distinguishing. But
nothing in Gangopadhyay teaches or suggests this, providing only a mention
of color that would not appear to be category specific. Nonetheless,
Appellant finds Li to be curative of this deficiency.
Appellant argues that Li expressly discloses analyzing certain useful
features depending on the category of an object, and cites the example
therein whereby the type of boat may be determined based on the color of
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

16
the boat (App. Br. 23). Respondent argues that all Li actually provides for is
distinguishing boats from one another by color, which is not a category
specific feature (Resp. Br. 10). We generally agree with Respondent.
While Appellant suggest that Li provides for “boat recognition and
characterization” (App. Br. 23), we find full support for the former and no
support for the latter (FF4). Li emphasizes context, such as water and sky,
in determining the presence of a boat, and also looks for specific
characteristics of specific types of boats (FF4). However, the clear intent of
Li is that “[t]his simple approach works well for finding possible boats” (Li,
p. 4). This is equally true for the later examples of recognizing vehicles and
buildings, where the focus is on identifying an object, as a boat, vehicle or
building, and not on distinguishing the particular object from others within
the same categories. In other words, the point of Li appears to be that a boat
can be discerned from an image, and that knowledge of different types of
boats aids in that process, but not that different types of boats are
characterized. We do not find that Li teaches or suggest more.
Thus, we do not find that one of ordinary skill in the art would have
been motivated to identify and apply category-specific features to
distinguish the determined objects, per claims 1 and 15, based on the
disclosures of Gangopadhyay and Li. Thus, with respect to the Examiner’s
decision to not adopt the rejection of claims 1 and 15 over Gangopadhyay
and Li, we do not find that the Examiner erred in that non-adoption.

Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

17
Dependent claims 2-4, 6-11, 13, 17-19, 21, and 22
We acknowledge that Appellant specifically argues that dependent
claims 2-4, 6-11, 13, 17-19, 21, and 22 should be rejected over the
combination of Gangopadhyay, Li, and Liu (App. Br. 24-26), but concur
with the Examiner’s non-adoption of their rejection for the same reasons
provided supra with respect to the non-adoption of the rejection of
independent claims 1 and 15.

ISSUE C
Ritter, Li, and Liu, Independent Claims 1 and 15
Appellant argues that the basis for not adopting the proposed
rejections involving Ritter are not clear from the RAN (App. Br. 26), and
that these rejections should be adopted for at least the same reasons as
discussed above with respect to the rejections involving Li (App. Br. 27).
Respondent continues to find Li to be lacking the necessary disclosure and
agrees with the Examiner that the addition of Ritter does not change the
Examiner’s decision to not adopt the rejections involving Ritter (Resp. Br.
10-11).
Based on our analysis of Li supra, we concur with the Examiner’s
decision to not adopt the rejections involving Ritter. Appellant reargues the
teachings of Li with respect to these proffered rejections, and we do not find
those arguments to be more convincing in the presence of Ritter. As such,
we find no error in the Examiner’s non-adoption of the rejection of claims 1
and 15 over Ritter, Li, and Liu.

Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

18
Dependent claims 2, 5-11, and 16-25
We acknowledge that Appellant specifically argues that dependent
claims 2, 5-11, and 16-25 should be rejected over the combination of Ritter,
Li, and Liu (App. Br. 27-28), but concur with the Examiner’s non-adoption
of their rejection for the same reasons provided supra with respect to the
non-adoption of the rejection of independent claims 1 and 15.

ISSUE D
Wang, Higashide, and Liu, Independent Claims 1 and 15
Appellant argues that the Examiner erred with respect to the findings
about Wang because Wang classifies an image into a semantic class and
then extracts features from the image based on the semantic type (App. Br.
29). Appellant argues that Wang refers to “object composition” as part of
the image semantics and bases its classification on the object in the image
(id.). Respondent argues that it is clear from Wang that the semantic
classification is applied to the entire image (Resp. Br. 11-12). We agree
with Respondent.
Appellant argues that “[r]egardless of whether a separate image
analysis step is identified for extracting a product from the image, the
classification is based on the object in the image, and not the entire image”
(Rebut. Br. 7). However, we find that to be the point. Claim 1, for example,
recites “(ii) determining a category of each identified object.” If an image,
in Wang, contained a single object, then the classification of the object
might occur, but this would be incidental to the classification of the image
overall. The disclosure of Wang equally allows for an image to be classified
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

19
as a “photograph,” which would cause no objects to be classified (FF5). We
concur with the Examiner’s judgment that “Wang does not appear to
disclose categorizing different identified objects within an image” (RAN 7).
Appellant also argues that the Examiner erred in not finding that
Higashide discloses performing image analysis on product images (App. Br.
29). Respondent argues that even if Higashide performed image analysis on
product images, its disclosure would not cure the deficiencies of Wang
(Resp. Br. 12). We agree with Respondent.
Appellant argues that Higashide “discloses (at least inherently)
obtaining or inferring a product from an image and performing image
analysis on the product” (Rebut. Br. 8). However, this does not disclose
“determining a category of each identified object,” as the system in
Higashide may disclose identification without categorization. In other
words, the multiple candidates for products inferred from the images can be
obtained without the use of categories (FF6). Even if Appellant is correct
that a manageable number of candidates would have to be reduced through
use of a “known taxonomic or hierarchical searching methodology” (App.
Br. 29), Higashide is silent with respect to such a methodology. As such, we
do not find that one of ordinary skill in the art would have been motivated to
determine a category for each identified object in view of Wang and
Higashide.
Appellant also argues the aspect of claim 1 not found in Wang by the
Examiner has been admitted by the patentee to be prior art in the '610 Patent
(App. Br. 29-30). Respondent argues that there has been no admission and
that the existing techniques refer to those described in a parent patent
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

20
reference incorporated by reference (Resp. Br. 12). Appellant responds that
the cited section of the '610 Patent references that only some of them are
stated in the incorporated reference and that others encompass prior art
techniques (Reply Br. 8). We agree with Respondent.
The section of the '610 Patent referred to is col. 6, ll. 1-5, which
begins: “[n]umerous techniques exist to determine objects in images.” We
do not, however, take this to be an admission that those techniques are all
prior art. The section does not make clear whether all of those techniques
could be applicable to the processes claimed in the '610 Patent, and does not
refer to such techniques as being “known,” only that they exist.
Nonetheless, the proffered rejection of claims 1 and 15 was over Wang and
Higashide, not Wang, Higashide, and the acknowledged prior art, such that
any further consideration would venture into a new ground not considered
by the Examiner.

Dependent claims 2-14 and 16-25
We acknowledge that Appellant specifically argues that dependent
claims 2-14 and 16-25 should be rejected over the combination of Wang,
Higashide, and Liu (App. Br. 31-33), but concur with the Examiner’s non-
adoption of their rejection for the same reasons provided supra with respect
to the non-adoption of the rejection of independent claims 1 and 15.
Appellant also argues that even if, assuming arguendo, Wang,
Higashide, and the alleged admitted prior art do not suggest the identifying
one or more objects in individual image content items, this element is
disclosed by Liu (App. Br. 30). We address this argument here because the
Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

21
combination of Wang, Higashide, and Liu was only proffered against claims
13, 14, and 17-20. We continue to find, as discussed supra, that Liu does
not disclose category-specific features, such that even if Liu would have
guided one of ordinary skill in the art to determine objects from an image,
neither Wang nor Higashide provide for the determination and use of
category-specific features as required by claims 1 and 15, from which claims
13, 14, and 17-20 depend. As such, we concur with the Examiner’s decision
to not adopt the proffered rejection of these claims.

CONCLUSIONS
We conclude that the Examiner did not err in not adopting A) the
rejection of claims 1, 2, 4-7, and 11-25 under 35 U.S.C. § 103(a) as being
unpatentable over Liu and Gokturk ‘906, B) the rejection of claims 1-4, 6-
11, 13, 15, 17-19, 21, and 22 under 35 U.S.C. § 103(a) as being unpatentable
over Gangopadhyay, Li, and Liu, C) the rejection of claims 1, 2, 5-11, and
15-25 under 35 U.S.C. § 103(a) as being unpatentable over Ritter, Li, and
Liu, and D) the rejection of claims 1-25 under 35 U.S.C. § 103(a) as being
unpatentable over Wang, Higashide, and Liu.

DECISION
The Examiner’s decision to not adopt the proffered rejections of
claims 1-25 is affirmed.
Requests for extensions of time in this inter partes reexamination
proceeding are governed by 37 C.F.R. § 1.956. See 37 C.F.R. § 41.79.

Appeal 2011-006145
Reexamination Control No. 95/001,307
Patent 7,542,610 B2

22
AFFIRMED

ack
cc:
FOR PATENT OWNER:
MAHAMEDI PARADICE KREISMAN LLP
550 WINCHESTER BOULEVARD
SUITE 605
SAN JOSE, CA 95128

FOR THIRD PARTY REQUESTER:

PEARL COHEN ZEDEK LATZER, LLP
1500 BROADWAY
12th FLOOR
NEW YORK, NY 10036