Ex Parte Lin et alDownload PDFPatent Trial and Appeal BoardJun 29, 201612869460 (P.T.A.B. Jun. 29, 2016) Copy Citation UNITED STA TES p A TENT AND TRADEMARK OFFICE APPLICATION NO. FILING DATE 12/869,460 108982 7590 Wolfe-SBMC 601 W. Main Avenue Suite 1300 Spokane, WA 99201 08/26/2010 07/01/2016 FIRST NAMED INVENTOR ZheLin 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. Bl228 7915 EXAMINER TRUONG, DENNIS ART UNIT PAPER NUMBER 2169 NOTIFICATION DATE DELIVERY MODE 07/01/2016 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): docket@sbmc-law.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte ZHE LIN and JONATHAN W. BRANDT Appeal2015-002859 1 Application 12/869,460 Technology Center 2100 Before JEAN R. HOMERE, JOHN A. EV ANS, and DANIEL J. GALLIGAN, Administrative Patent Judges. Per Curiam. DECISION ON APPEAL STATEMENT OF THE CASE Appellants seek our review under 35 U.S.C. § 134(a) from the Examiner's final rejection of claims 21--40. Claims 1-20 have been canceled. Br. 5. We have jurisdiction under 35 U.S.C. § 6(b). We affirm. 1 Appellants identify the real party in interest as Adobe Systems Incorporated. Br. 3. Appeal2015-002859 Application 12/869,460 Appellants ; Invention Appellants' invention is directed to a method and system for image retrieval and characterization involving a query image 400 compared with retrieval results 415, 420 from an image database. Spec. i-fi-182-83; Fig. 4. Illustrative Claim Independent claim 21 is illustrative, and reads as follows: 21. A method implemented by one or more computer systems, the method comprising: determining whether a query image corresponds to one or more of a set of images by: determining whether one or more features of the query image correspond to one or more features in respective images in the set of images, the one or more features defining one or more objects in the respective image; and determining whether spatial locations of the one or more features of the query image are spatially consistent with spatial locations of the one or more features of the respective images in the set of images, the spatial locations describing positioning of respective said features within a respective said image; and returning a result of the determining of whether the query image corresponds to one or more of the set of images. Prior Art Relied Upon Svetlana Lazebnik et al., Beyond Bags of Features: Spatial Pyramid Matching/or Recognizing Natural Scene Categories, 2006 IEEE Conference on Computer Vision and Pattern Recognition (CVPR'06) 2169-78 (2006) ("Lazebnik"). 2 Appeal2015-002859 Application 12/869,460 Yang Cao et al., Spatial-Bag-of-Features, 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 3352-59 (June 18, 2010) ("Cao"). Re} ection on Appeal Appellants request review of the Examiner's rejection of claims 21- 40 under 35 U.S.C. § 103(a) as being unpatentable over Cao and Lazebnik. ANALYSIS We consider Appellants' arguments seriatim, as they are presented in the Appeal Brief, pages 11-23. 2 Dispositive Issue: Under 35 U.S.C. § 103, did the Examiner err in finding the combination of Cao and Lazebnik teaches or suggests the one or more features defining one or more objects in the respective image and determining whether spatial locations of the one or more features of the query image are spatially consistent with spatial locations of the one or more features of the respective images in the set of images, the spatial locations describing positioning of respective said features within a respective said image, as recited in independent claim 21? Appellants first argue the Lazebnik does not teach the disputed limitations emphasized above. Br. 12-13. In particular, Appellants assert Lazebnik teaches recognizing scene categories based on the partitioning of 2 Rather than reiterate the arguments of Appellants and the Examiner, we refer to the Appeal Brief (filed July 30, 2014), and the Answer (mailed October 23, 2014) for the respective details. We have considered in this Decision only those arguments Appellants actually raised in the Brief. Any other arguments Appellants could have made but chose not to make in the Brief are deemed to be waived. See 37 C.F.R. § 41.37(c)(l)(iv) (2012). 3 Appeal2015-002859 Application 12/869,460 an image into increasingly fine sub-regions. Br. 12 (citing Lazebnik, Abstract, § 1 i-f 2). Appellants further contend Lazebnik uses histograms of local features found inside each sub-region to compute a rough geometric correspondence of an entire image (i.e., the "gist" of the image) to provide information for a subsequent search of objects. Id. at 12-13 (citing Lazebnik, Abstract; § 1 ). Appellants moreover argue Lazebnik teaches features generally to categorize a scene rather than "features defining one or more objects in the respective image," as recited in the claim. Id. at 13. These arguments are not persuasive. At the outset, we note Appellants' arguments referenced hereinabove constitute an individual attack against the references. 3 Appellants' arguments additionally attack each reference for deficient teachings for which the Examiner relies upon the other reference to provide. Specifically, the Examiner relies on Cao, not Lazebnik, for its teaching of features defining one or more objects in the respective image. Final Act. 4 (citing Cao§ 3.1, Eqs. 3, 4). The Examiner relies on Lazebnik, not Cao, to teach the determining step. Ans. 2-3 (citing Lazebnik §§ 4, 3.1, 3.2). The Examiner finds Lazebnik teaches weak and strong features, where weak features are used to determine the "gist" of the image and strong features provide a determination of objects. Id. at 2 (citing Lazebnik § 4). The Examiner further finds Lazebnik teaches using the histogram measurements of features in the same location of the image to determine features of an object in that same location. Id. at 2-3 (citing Lazebnik §§ 3.1, 3.2). We 3 See In re Keller, 642 F.2d 413, 426 (CCPA 1981) ("[O]ne cannot show non-obviousness by attacking references individually where, as here, the rejections are based on combinations of references"). 4 Appeal2015-002859 Application 12/869,460 agree with the Examiner that Lazebnik teaches a two-stage process with the first stage, identified by Appellants (Br. 12-13), relying upon weak features to provide information used in combination with strong features in a second stage to teach the determining step. Appellants fail to rebut the Examiner's findings regarding the second stage of this process. Appellants next argue Lazebnik cannot be combined with Cao because it teaches away from object recognition. Br. 13-14. In particular, Appellants assert Lazebnik does "not advocate the direct use of a global method for object recognition." Id. at 12, 14 (quoting Lazebnik § 1). Appellants further contend Lazebnik frustrates the purpose of Cao "to encode geometric information of objects within an image." Id. at 14 (quoting Cao, Abstract). As explained above, Appellants argue the "global method" in Lazebnik corresponds to the first stage in a two-stage process. Lazebnik § 4. The second stage with the use of the strong features "is necessary to capture uniform regions such as sky, calm water, or road surface." Id. Contrary to Appellants' assertion, the implementation of strong features in Lazebnik is consistent with Appellants' stated purpose of Cao because the strong features represent encoded geometric information. Appellants contend Cao teaches away from incorporating Lazebnik. Br. 14. In particular, Appellants assert Cao states the linear ordered bag-of- features it calculates can "capture other slantwise directions that [spatial pyramid matching (SPM) taught in Lazebnik] cannot handle." Id. at 14-- 15 (quoting Cao § 2 .2). Appellants argue the determination of slantwise directions in Cao is an improvement over Lazebnik to discourage its combination with Cao. Id. at 15. The Examiner finds the ordered bag-of- features taught in Cao is similar to the spatial pyramid matching in 5 Appeal2015-002859 Application 12/869,460 Lazebnik. Ans. 2 (citing Cao § 1 ). The Examiner finds Cao expands upon spatial pyramid matching techniques to account for object translation, rotation, and scaling. Id. In this way, we obtain a kind of so-called ordered bag-of- features. This is a generalization of the spatial pyramid matching idea[.] However, in terms of utilizing of spatial constraints, these features are too rigorous to handle typical transformations of objects, i.e., translation, rotation, scaling. Therefore, we further process the ordered bag-of-features to obtain a kind of so-called spatial bag-of-features by some operations for histogram features, i.e. calibration, equalization and decomposition. Cao § 1. Cao explains the ordered bag-of-features, like spatial pyramid matching, is limited to determining object translation, rotation, and scaling. Id. Cao teaches further processing to determine these characteristics is performed through spatial bag-of-features. Therefore, we agree with the Examiner that Cao expands upon, with further processing, spatial pyramid matching to determine qualities, such as a slantwise direction of an object in the event the object is rotated. Ans. 2 (citing Cao § 1 ). Accordingly, Appellants have not persuaded us of error in the Examiner's rejection of claim 21. Regarding claim 29, Appellants argue the combination of Cao and Lazebnik does not teach constructing an inverted file indexed by the visual words, each of the visual words having a corresponding indication of location of the local interest region within the image. Br. 16-17. Specifically, Appellants contend Lazebnik does not teach visual words or an inverted file indexed by the visual words. Id. at 17 (citing Lazebnik § 3.2). However, the Examiner relies on Cao, not Lazebnik, to teach an inverted file 6 Appeal2015-002859 Application 12/869,460 indexed by visual words. Ans. 3 (citing Cao § 3 .1 ). The Examiner explains H in Cao denotes a histogram representing the frequency of a word v in a bin i. Id. We agree with the Examiner that this teaching represents an inverted file indexed by visual words. Id. The Examiner relies on Lazebnik to teach an indication of location of the local interest region within the image. Ans. 3--4 (citing Lazebnik § 3.2); see Lazebnik § 3.2 ("Each channel m gives us two sets of two-dimensional vectors, Xm and Ym, representing the coordinates of features of type m found in the respective images."). The channel m in Lazebnik, like the word v in Cao, represents a feature. See Lazebnik § 3 .2 ("we quantize all feature vectors into [ m] discrete types."); see Cao § 3 .1 ("each feature v"). Thus, we agree with the Examiner that Lazebnik teaches using Xm and Ym to determine the location of a feature, m, designated by word v in Cao, in the image. Ans. 4 (citing Lazebnik § 3 .2 ); see Cao § 3 .1. It follows Appellants have not shown error in the Examiner's rejection of claim 29. Regarding the rejections of claims 22-28 and 30--40, because Appellants have not presented separate persuasive patentability arguments or have reiterated substantially the same arguments as those previously discussed for the patentability of claims 21 and 29 above (Br. 17-23), claims 22-28 and 30--40 fall therewith. See 37 C.F.R. § 41.37(c)(l)(iv). DECISION We affirm the Examiner's obviousness rejection under 35 U.S.C. § 103(a) of claims 21--40. 7 Appeal2015-002859 Application 12/869,460 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 8 Copy with citationCopy as parenthetical citation