Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
This Final Rejection is filed in response to Applicant Arguments/Remarks made in an Amendment filed 02/20/2026.
Claims 1, 12, and 17 are amended.
Claims 1-20 remain pending.
Response to Arguments
Argument 1, Applicant argues in Applicant Arguments/Remarks made in an Amendment filed 02/20/2026 pg. 8-12, that prior art Ning fails to teach the Claim 1 limitation, “a user input generated via the input portion as a user selecting representations of at least two of the plurality of objects”.
Response to Argument 1, the examiner respectfully disagrees. Ning teaches a user interface for a user to select an outfit by adding one or garments to their cart, and subsequently running a compatibility analysis to determining an additional item to be added to the outfit based on the compatibility score.
This is supported by the following paragraphs of Ning
para. [0061, 0100], “in response to a user showing his/her interest in an item/garment (a target item/garment) on an e-commerce platform loaded in his/her terminal device 110, for example, adding the item/garment to his/her cart… the item recommendation device 200 may determine one or more fashion items to be recommended for being collocated with the target item/garment to form a recommended outfit together with the same… the compatibility value/score between element descriptors may represent the overall compatibility between items/garments within a candidate outfit. It may explicitly show how much an element is compatible with other elements ”.
Thus the BRI for “a user input generated via the input portion as a user selecting representations of at least two of the plurality of objects”, encompasses how a user can select one or more items of clothing before a compatibility score is calculated with other items of clothing to be recommended. Para. [0100], further augments this as for a whole outfit a compatibility score is calculated between each element. So for two user selected items of clothing a compatibility score is calculated. Ning also teaches a user inputting one or more item/garments. A user of Ning could add at least two garments and the system of Ning would calculate the overall compatibility between items/garments within a candidate outfit.
Argument 2, Applicant argues in Applicant Arguments/Remarks made in an Amendment filed 02/20/2026 pg. 12-15, that prior art Ning fails to teach the Claim 12 limitation, “the inputs selecting a first said attribute of a first said object and a second said attribute of a second said object, and specifying an affinity of the first said attribute and the second said attribute, respective attributes of the plurality of objects, one to another”.
Response to Argument 2, Applicant’s argument have been considered, however in light of the amendments a newly found combination of prior art (U.S. Patent Application Publication No. 20210035187 “Ning”, further in light of U.S. Patent Application Publication No. 20180308149 “Guo”, and further in light of Scoles, S. (2020, May 26). This Citizen Science Gig Pays People to Match Space Photos. WIRED. https://www.wired.com/story/this-citizen-science-gig-pays-people-to-match-space-photos/, hereinafter “Scoles”) is applied to updated rejections.
Argument 3, Applicant argues in Applicant Arguments/Remarks made in an Amendment filed 02/20/2026 pg. 15-17, that prior art Ning fails to teach the Claim 17 limitation, “generating training data based on the identifying, the training data correlating the plurality of attributes and objects as included based on inclusion of the plurality of attributes and objects together in respective said digital images”.
Response to Argument 2, the examiner respectfully disagrees. Ning teaches the generation of training data for intake of a CNN model that outputs a compatibility score between elements in a candidate fashion outfit. The model is trained on training data is generated by expert labeling the training data with weight thresholds defining the compatibility of elements forming an output.
This is supported by Ning para. [0096], “the retained fashion elements may be sequentially fed to the element analyzer 246, and an attention mechanism for weight calculation of elements may be employed in order to give importance to those elements which hold more significance (or feature elements)… In some embodiments, the attention mechanism may also be implemented by a CNN model, which may be trained in advance using training data labeled according to experts' advices and/or suggestions”.
The BRI for “the plurality of attributes and objects together in respective said digital images”, encompasses how training data for an outfit has a plurality of items of clothing, each of which may be generated to have an assigned significance attribute in their analyzed images. The examiner further notes that the limitation “generating” is broad enough to encompasses the labeling of training data by experts.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 17-20 is/are rejected under 35 U.S.C. 102(a)(2) as being anticipated by U.S. Patent Application Publication No. 20210035187 “Ning”.
Claim 17:
Ning teaches a computing device comprising:
a processing device (i.e. para. [0066], The processor 212 may be a central processing unit (CPU) which is configured to control operation of the computing device 200); and a computer-readable storage medium storing instructions that, responsive to execution by the processing device (i.e. para. [0055], The computer programs include processor-executable instructions that are stored on a non-transitory tangible computer readable medium), causes the processing device to perform operations including:
receiving a plurality of training digital images (i.e. para. [0106], via the interface provided by the interface module 270, the manager may issue an instruction to update the raw image database 290, the training database 294); identifying a plurality of objects and a plurality of attributes included in respective objects of the plurality of objects in the plurality of training digital images (i.e. para. [0077], Also, each image in the set of training data may be partitioned into a plurality of (for example, n, which should be less than or equal to N) human related regions according to the predefined categories); generating training data based on the identifying, the training data correlating the plurality of attributes and objects as included based on inclusion of the plurality of attributes and objects together in respective said digital images (i.e. para. [0096], “In some embodiments, the attention mechanism may also be implemented by a CNN model, which may be trained in advance using training data labeled according to experts' advices and/or suggestions”, wherein the BRI for “based on inclusion of the plurality of attributes and objects together in respective said digital images”, encompasses how training data for an outfit has a plurality of items of clothing, each of which may be generated to have an assigned significance attribute in their analyzed images); training a machine-learning model (i.e. para. [0078], the fashion keypoint/fashion landmark aided human parsing module 232 may be trained in advance using a set of training data included in the training data 294, where each image in the set of training data may be labeled with their corresponding fashion feature) to generate an affinity score using the training data, the affinity score quantifying an amount of affinity respective said attributes have to each other of respective said objects (i.e. para. [0100], the compatibility determining module 260 which, when loaded into the memory 214 and executed by the processor 212, may cause the processor 212 to calculate a compatibility value/score between element descriptors for a set of items/garments forming a candidate outfit); and outputting the trained machine-learning model (i.e. para. [0101], the compatibility score calculating module 262 which, when loaded into the memory 214 and executed by the processor 212, may cause the processor 212 to, after receiving the element descriptors, calculate a compatibility value/score between the element descriptors for the set of items forming the candidate outfit. In some embodiments, the compatibility score calculating module 262 may be implemented by a convolutional neural network (CNN)) .
Claim 18:
Ning teaches the computing device as described in claim 17, wherein the identifying the plurality of objects and the plurality of attributes is performed automatically and without user intervention using a machine-learning model (i.e. para. [0058], “the item recommendation device 200 may determine one or more fashion items (e.g. garments, accessories, etc.) to be recommended for being collocated with a target item/garment (or an item/garment of interest) to form a recommended outfit together with the same, and output information of the recommended outfit”, wherein the at least two recommended garments to be included as determined to have sufficiently high enough compatibility scores with each other is a separate and automatic input in response to the user input of a first item. As such the user has no need to specify another input or intervention in this embodiment where a whole outfit may be recommended to a user).
Claim 19:
Ning teaches the computing device as described in claim 17, wherein the generating the training data includes positive training samples based on the plurality of training digital images and negative training samples generated by editing one or more of the plurality of training digital images (i.e. para. [0096], “In some embodiments, the attention mechanism may also be implemented by a CNN model, which may be trained in advance using training data labeled according to experts' advices and/or suggestions. After the weight calculation of elements, an element having a relatively low weight, for example, lower than a predefined weight threshold, may be discarded, while an element having a relatively high weight, for example, higher than the predefined weight threshold, may be retained. The retained fashion elements are the so-called feature elements that define the outfit style and the overall compatibility of a candidate outfit. As a result, feature elements are determined from among fashion elements of a set of items forming a candidate outfit.”, wherein the BRI for positive and negative training samples encompasses how the images may have their weight edited to be above or below a weight threshold for retention).
Claim 20:
Ning teach the computing device as described in claim 17, wherein the plurality of attributes includes style (i.e. para. [0096], The retained fashion elements are the so-called feature elements that define the outfit style and the overall compatibility of a candidate outfit) or color (i.e. para. [0092], The fashion elements may be comprehensive features such as color, pattern, texture, and physical properties).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-3 & 6-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No. 20210035187 “Ning” and further in light of U.S. Patent Application Publication No. 20200134694 “Park”.
Claim 1:
Ning teaches a method comprising:
displaying, by a processing device (i.e. para. [0106], the interface module 270 which, when loaded into the memory 214 and executed by the processor 212),
a user interface having an input portion and a plurality of representations of a plurality of objects (i.e. para. [0061], “in response to a user showing his/her interest in an item/garment (a target item/garment) on an e-commerce platform loaded in his/her terminal device 110, for example, adding the item/garment to his/her cart on the e-commerce platform and/or purchasing the item/garment on the e-commerce platform etc., the item recommendation device 200 may determine one or more fashion items to be recommended for being collocated with the target item/garment to form a recommended outfit together with the same, and then send information of the recommended outfit, for example, an outfit image, brands, inventories, prices, and/or sizes of respective recommended fashion items/garments within the outfit, and purchasing information, etc., to the user via the network 120 and display the same on a display unit”, wherein the user interface may have an ecommerce input portion with a plurality of representations of clothing objects)
receiving, by the processing device, a user input generated via the input portion as a user selecting representations of at least two of the plurality of objects (i.e. para. [0061], “the item recommendation device 200 may determine one or more fashion items to be recommended for being collocated with the target item/garment to form a recommended outfit together with the same, and then send information of the recommended outfit”, wherein in an embodiment the processor receives one or more, which may be two, of the plurality of new outfit items as candidate outfits to be evaluative for compatibility without user input as the user only inputs a first outfit item by adding or purchasing the first item. Wherein the BRI for “a user input generated via the input portion as a user selecting representations of at least two of the plurality of objects”, encompasses how a user can select one or more items of clothing before a compatibility score is calculated with other items of clothing to be recommended);
obtaining, by the processing device, an affinity score based on the at least two objects, the affinity score specifying a relative amount of affinity of the at least two objects, one to another (i.e. para. [0100], “the compatibility value/score between element descriptors may represent the overall compatibility between items/garments within a candidate outfit. It may explicitly show how much an element is compatible with other elements”, wherein the BRI for an affinity score encompasses a compatibility score between each garment in the potential outfit set and how each garment object has a compatibility score specified to each other); and
displaying, by the processing device, the affinity (i.e. para. [0100], It may explicitly show how much an element is compatible with other elements; or, how much an item/garment (represented by a set of elements) is compatible with other items/garments (also represented by a set of elements). Thus, the compatibility determining module 260 is used to determine whether items within a candidate outfit are compatible with each other, or whether compatibility value/score of a candidate outfit is sufficiently high so that it can be a recommended outfit”, wherein the compatibility determining module 260 may display an affinity how much a garment object is compatible with other garments as the display of garment outfits that have a sufficiently high enough compatibility score).
While Ning teaches displaying the affinity in the portion of the user interface by displaying representations of the at least two objects that have a sufficiently high compatibility score, Ning may not explicitly teach displaying
the affinity score in the portion of the user interface along with the representations of the at least two objects.
However Park teaches
the affinity score in the portion of the user interface along with the representations of the at least two objects (i.e. para. [0119], Fig. 6, “, the value of the similarity of a similar item, and the value of the similarity of a recommended item or computing the mean value thereof, and displays the recommended fashion outfit sorted in descending order to the user, it is possible to also display the fit score of each set of recommended fashion outfit 460, as shown in the lower end of FIG. 6.”, wherein the BRI for an affinity score encompasses the display of a fit score under an outfit set).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention the display of an affinity score, with Ning’s machine learning model for user affinity recommendations, and displaying an affinity score in the portion of the user interface along with the representations of the at least two objects, as taught by Park. One would have been motivated to combine display of a calculated outfit score of Park with the display of recommended outfits of Ning in order to present fashion outfit such that the user easily evaluates and purchases the fashion items.
Claim 2:
Ning and Park teach the method as described in claim 1.
Ning further teaches wherein the affinity score is based on a relative amount of affinity that attributes, of the respective at least two objects, have to each other (i.e. para. [0100], “the compatibility determining module 260 is used to determine whether items within a candidate outfit are compatible with each other, or whether compatibility value/score of a candidate outfit is sufficiently high so that it can be a recommended outfit. As illustrated in FIG. 2D, the compatibility determining module 260 may include, among other things, a compatibility score calculating module 262 and a compatibility score comparing module 264”, wherein it is noted that the compatibility score may be based on affinity that extracted attribute vectors of each garment have towards each other. Wherein the BRI for attributes encompasses any data that describes a garment).
Claim 3:
Ning and Park teach the method as described in claim 1.
Ning further teaches
wherein the receiving the input includes specifying inclusion of the representations of the at least two objects in the input portion and the obtaining is performed automatically and without user intervention responsive to the input (i.e. para. [0058], “the item recommendation device 200 may determine one or more fashion items (e.g. garments, accessories, etc.) to be recommended for being collocated with a target item/garment (or an item/garment of interest) to form a recommended outfit together with the same, and output information of the recommended outfit”, wherein the at least two recommended garments to be included as determined to have sufficiently high enough compatibility scores with each other is a separate and automatic input in response to the user input of a first item. As such the user has no need to specify another input or intervention in this embodiment where a whole outfit may be recommended to a user).
Claim 6:
Ning and Park teach the method as described in claim 1.
Ning further teaches
wherein the obtaining is configured to cause generation of the affinity score (i.e. para. [0101], he compatibility score calculating module 262 may be implemented by a convolutional neural network (CNN)), automatically and without user intervention, using a machine-learning model (i.e. para. [0096], the attention mechanism may also be implemented by a CNN model, which may be trained in advance using training data labeled according to experts' advices and/or suggestion”, wherein since the model is trained in advance, the deployed model operates without user intervention).
Claim 7:
Ning and Park teach the method as described in claim 6
Ning further teaches
wherein the machine-learning model is trained by:
selecting a training digital image from a plurality of training digital images (i.e. para. [0099], the element descriptor generating module 250 may be pre-trained on datasets. Specifically, the element descriptor generating module 250 may be trained in advance using a set of training data included in the training data 294); generating object and attribute data based on the selected training digital image, the object and attribute data describing a correlation of objects as identified within the selected training digital image and respective attributes of the objects included within the selected training digital image (i.e. para. [0099], the feature elements may be fed to the element descriptor generating module 250, so as to generate an element descriptor of the item based on the feature elements. The element descriptor of an item may be a comprehensive feature representation of the item, and contain substantially all the description information of the item, including information about various feature elements thereof); and training the machine-learning model based on the object and attribute data for the plurality of training digital images (i.e. para. [0109], the training database 294 may include data for training various modules in the computing device 200. Each set of data in the training data 294 may correspond to a specific module, and be labeled with corresponding features. For example, a set of data may be labeled with corresponding segmentation maps, and the set of data may be used to train the fashion element extracting module 24).
Claim 8:
Ning and Park teach the method as described in claim 7.
Ning further teaches further comprising
identifying the objects within the selected training digital image and the respective attributes of the objects using one or more machine learning models, automatically and without user intervention, and wherein the generating the object and attribute data is based on the identifying (i.e. para. [0098], “the element descriptor generating module 250 which, when loaded into the memory 214 and executed by the processor 212, may cause the processor 212 to, upon receiving the determined feature elements, generate an element descriptor based on the feature elements. In some embodiments, the element descriptor generating module 250 may be implemented by a convolutional neural network (CNN). The CNN is configured to convert the feature areas into a feature vector having a fixed length. In other words, the CNN extract features from the feature elements”, wherein the model may be pre-trained and then deployed to run without user intervention from memory).
Claim 9:
Ning and Park teach the method as described in claim 1.
Ning further teaches
wherein the generating the affinity score is based on one or more affinity rules, the one or more affinity rules specifying a relative amount of affinity of attributes of the at least two objects have to each other (i.e. para. [0129], the compatibility score calculating module 262, after receiving the element descriptors, calculates a compatibility value/score between the element descriptors for the set of items forming the candidate outfit …. In some embodiments, a blacklist (a pair of items/garments that are incompatible) and a white list (a pair of items/garments that are compatible) may be provided by fashion professionals. When a pair of items in an outfit is listed in the blacklist, there is no need to extract fashion elements, and the outfit is regarded as a bad combination of items and the outfit will not be recommended to the customers).
Claim(s) 4-5 is/are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No. 20210035187 “Ning” and further in light of U.S. Patent Application Publication No. 20200134694 “Park”, as applied to claim 1 above, and further in light of U.S. Patent Application Publication No. 20100306066 “Binnewies”.
Claim 4:
Ning and Park teach the method as described in claim 1.
Ning and Park may not explicitly teach
wherein the user interface is configured to persist display of the input portion during navigation between a plurality of webpages.
However, Binnewies teaches
wherein the user interface is configured to persist display of the input portion during navigation between a plurality of webpages (i.e. para. [0011], “A first advertisement may be loaded and displayed along with the first web page. Ad-persist information is stored regarding the first advertisement. When the user navigates to a second web page, the first web page and the first advertisement are unloaded. The second web page is then loaded. A determination is made, based upon the ad-persist information, whether the first advertisement is to be displayed for the second web page”, wherein the ad-persist space is display space on a web page that maintains display space as a user navigates along multiple web pages).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to add wherein the user interface is configured to persist display of the input portion during navigation between a plurality of webpages, with Ning-Park’s machine learning model for user affinity recommendations, and a user interface is configured to persist the display of recommended advertisements across multiple web pages, as taught by Binnewies. One would have been motivated to combine persistent display of ads of Binnewies with the display of recommended outfits of Ning-Park as both are in the similar art of providing recommendations and adding a persistent space for recommended information can provide a user with enough time to see and/or engage the individual ads.
Claim 5:
Ning, Park, and Binnewies teach the method as described in claim 4.
Binnewies further teaches
wherein the plurality of representations of the plurality of objects is changed responsive to the navigation between the plurality of webpages (i.e. para. [0012], a request may be communicated to an advertisement server for another advertisement. A new advertisement served in response to the request may then be loaded for the second web page. The ad-persist information may be updated to store information related to the new advertisement).
Claim(s) 10-11 is/are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No. 20210035187 “Ning” and further in light of U.S. Patent Application Publication No. 20200134694 “Park”, as applied to Claim 9 above, and further in light of U.S. Patent Application Publication No. 20180308149 “Guo”.
Claim 10:
Ning and Park teach the method as described in claim 9.
Ning further teaches wherein the one or more affinity rules are generated by:
(i.e. para. [0129], In some embodiments, a blacklist (a pair of items/garments that are incompatible) and a white list (a pair of items/garments that are compatible) may be provided by fashion professionals).
While Ning teaches affinity rules based on professional input, Ning may not explicitly teach wherein the one or more affinity rules are generated by:
outputting a rule generation user interface including a plurality of representations of a plurality of attributes, respectively, of a plurality of objects; receiving inputs via the rule generation user interface, the inputs specifying an affinity of respective said attributes of the plurality of objects.
However, Guo teaches
outputting a rule generation user interface including a plurality of representations of a plurality of attributes (i.e. para. [0054], “Examples of attributes and values are: Clothing Category (Top, Bottom, Setwear, Outerwear, Shoes, Bags, Scarfs); Top Type (Blouse, T-shirt, Button Shirt, Sweater, Tank Top, Cardigan, Camisole, Vest); Blouse Sleeve Length (Sleeveless, Cap Sleeve, Short Sleeve, Long Sleeve)”, wherein it is noted that the each outfit garment may have a plurality of attributes), respectively, of a plurality of objects (i.e. para. [0068], “FIG. 5 depicts an outfit library 500 and the outfits 510 therein. Each outfit 510 in the outfit library 500 may be displayed for a user”, wherein it is noted that a user may generate rules for a recommendation engine by providing feedback on outfits); receiving inputs via the rule generation user interface, the inputs specifying an affinity of respective said attributes of the plurality of objects (i.e. para. [0068], a user may provide feedback on the outfits, such as by liking them, rejecting them, disliking them, selecting favorites, and other types of feedback. The feedback may be used by the recommendation engine).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to add wherein outputting a rule generation user interface including a plurality of representations of a plurality of attributes, respectively, of a plurality of objects; receiving inputs via the rule generation user interface, the inputs specifying an affinity of respective said attributes of the plurality of objects, with Ning-Park’s machine learning model for user affinity recommendations that may operate off affinity rules set by fashion professionals , and how a recommendation engine may operate off user preference and affinity rules set by a user themselves in an interface, as taught by Guo. One would have been motivated to combine Guo with Ning-Park as both are in the similar art of providing recommendations and that adding user set constraints to a recommendation model can provide better quality recommendations that are also aligned with user's personal style.
Claim 11:
Ning, Park, and Guo teach the method as described in claim 9.
Ning further teaches wherein the one or more affinity rules include:
a first said affinity rule specifying a positive affinity between a first said attribute of a first said object and a second said attribute of a second said object; and a second said affinity rule specifying a negative affinity between a third said attribute of a third said object and a fourth said attribute of a fourth said object (i.e. para. [0129], “In some embodiments, a blacklist (a pair of items/garments that are incompatible) and a white list (a pair of items/garments that are compatible) may be provided by fashion professionals”, wherein a first rule for white list may be a positive compatibility for a first garment and a second garment and a second rule for a blacklist may be a negative or effectively zero compatibility for a third garment and a fourth garment. The examiner notes that the claims do not require any of the attributes to be different or the same).
Claim(s) 12-16 is/are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication No. 20210035187 “Ning”, further in light of U.S. Patent Application Publication No. 20180308149 “Guo”, and further in light of Scoles, S. (2020, May 26). This Citizen Science Gig Pays People to Match Space Photos. WIRED. https://www.wired.com/story/this-citizen-science-gig-pays-people-to-match-space-photos/, hereinafter “Scoles”.
Claim 12:
Ning teaches one or more computer-readable storage media storing instructions that, responsive to execution by a processing device, causes the processing device to perform operations including:
outputting a rule generation user interface (i.e. para. [0129], a blacklist (a pair of items/garments that are incompatible) and a white list (a pair of items/garments that are compatible) may be provided by fashion professionals) (i.e. para. [0129], “the attention mechanism may also be implemented by a CNN model, which may be trained in advance using training data labeled according to experts' advices and/or suggestions”, wherein it is noted the BRI for inputs for respective attributes encompasses the respective weight calculations for outfits and their attributes as specified by the attention mechanism that was trained according to expert inputs and suggestions); generating one or more affinity rules based on the affinity of the respective attributes of the plurality of objects as specified by the inputs (i.e. para. [0129], When a pair of items in an outfit is listed in the blacklist, there is no need to extract fashion elements, and the outfit is regarded as a bad combination of items and the outfit will not be recommended to the customers); and generating an affinity score based on a plurality of subsequent objects using the one or more affinity rules (i.e. para. [0129], “the outfit is regarded as a bad combination of items and the outfit will not be recommended to the customers”, wherein an effective affinity score of zero is generated based on the blacklisted outfit combinations generated by an expert).
While Ning teaches a rule generation interface for experts to provide input rules on suggested and not suggested outfit combination affinities, Ning may not explicitly teach that the rule generation interface is
Including a plurality of representations of a plurality of attributes, respectively, of a plurality of objects .
However, Guo also teaches
A rule generation interface (i.e. para. [0068], a user may provide feedback on the outfits, such as by liking them, rejecting them, disliking them, selecting favorites, and other types of feedback. The feedback may be used by the recommendation engine).
Guo further teaches wherein the interface is
Including a plurality of representations of a plurality of attributes, respectively, of a plurality of objects (i.e. para. [0068], Fig. 5, FIG. 5 depicts an outfit library 500 and the outfits 510 therein. Each outfit 510 in the outfit library 500 may be displayed for a user)
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to add that a rule generation interface is including a plurality of representations of a plurality of attributes, respectively, of a plurality of objects, with Ning’s machine learning model for user affinity recommendations that may operate off affinity rules set by fashion professionals, and how a recommendation engine may operate off user preference and affinity rules set by a user themselves in an interface, as taught by Guo. One would have been motivated to combine Guo with Ning as both are in the similar art of providing recommendations and that adding user set constraints to a recommendation model can provide better quality recommendations that are also aligned with user's personal style.
While Ning-Guo teach an affinity rule generation interface with a plurality of objects with attributes, Ning-Guo may not explicitly teach that the rule generation interface recites inputs selecting
a first said attribute of a first said object and a second said attribute of a second said object, and specifying an affinity of the first said attribute and the second said attribute.
However, Scoles teaches
Selecting a first said attribute of a first said object and a second said attribute of a second said object, and specifying an affinity of the first said attribute and the second said attribute (i.e. pg. 8-9, “Obermaier and his new colleagues will be going through 2,000 images, chosen by White, comparing each one to all the others. That’s about 2 million pairs. Their task is to evaluate which pictures resemble each other, comparing them side by side. “Which ones are most similar, which ones are sort of similar but not very close, and which ones are not very similar?”… Looking at whether this noctilucent arm resembles that one, or whether two star sets look like fraternal twins or thrice-removed cousins”, wherein the BRI for a first and second attribute of a first and second object encompasses how a user may select how much galaxies present in two different photos are alike to each other).
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to add selecting a first said attribute of a first said object and a second said attribute of a second said object, and specifying an affinity of the first said attribute and the second said attribute, with Ning-Guo’s machine learning model for generating recommendations that may operate off affinity rules set by users specifying how well clothing objects and their characteristics match an outfit, and how a Zooniverse crowdsource website interface teaches how users may specify an affinity score that represents subjective matching classification between the attributes of object photos, as taught by Scoles. One would have been motivated to combine specific user interface for inputting an affinity score between two objects of Scoles and the clothing specific affinity scoring of two outfit garments of Ning-Guo as both are in the similar art user sourced classification and that adding user specified affinity interfaces to the ground work of an expert derived fashion rule list can provide an increase of crowd sourced data for a neural net as humans generally create the data sets that train the networks, curating a set of images that show lots of the objects that the net is learning to seek.
Claim 13:
Ning, Guo, and Scoles teach the one or more computer-readable storage media as described in claim 12.
Guo further teaches wherein the receiving of the inputs is performed responsive to user interaction with a control in the user interface (i.e. para. [0068], a user may provide feedback on the outfits, such as by liking them, rejecting them, disliking them, selecting favorites, and other types of feedback. The feedback may be used by the recommendation engine).
Claim 14:
Ning, Guo, and Scoles teach the one or more computer-readable storage media as described in claim 13.
Guo further teaches wherein the control is configurable to specify a relative amount of affinity between a first said attribute of a first said object and a second said attribute of a second said object (i.e. para. [0068], “a user may provide feedback on the outfits, such as by liking them, rejecting them, disliking them, selecting favorites, and other types of feedback. The feedback may be used by the recommendation engine”, wherein the BRI for a relative amount of affinity encompasses a like or a dislike of an outfit combination).
Claim 15:
Ning, Guo, and Scoles teach the one or more computer-readable storage media as described in claim 12.
Guo further teaches wherein at least one said affinity rule specifies a positive affinity between a first said attribute of a first said object and a second said attribute of a second said object (i.e. para. [0068], a user may provide feedback on the outfits, such as by liking them, rejecting them, disliking them, selecting favorites, and other types of feedback. The feedback may be used by the recommendation engine).
Claim 16:
Ning, Guo, and Scoles teach the one or more computer-readable storage media as described in claim 12.
Guo further teaches wherein at least one said affinity rule specifies a negative affinity between a first said attribute of a first said object and a second said attribute of a second said object (i.e. para. [0068], a user may provide feedback on the outfits, such as by liking them, rejecting them, disliking them, selecting favorites, and other types of feedback. The feedback may be used by the recommendation engine).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
U.S. Patent No. 11527032 “Baron” teaches in Col 10, lines 5-20, User interface elements 404a-c (e.g., virtual sliders) may enable a user to provide levels of correlation to content styles. The levels of correlation input by the user may adjust compiled animation scene 402 in real-time to generate the styled compiled animation scene. The styled compiled animation scene may be generated, stored, and/or transmitted upon selection of an export button 406. Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVID H TAN whose telephone number is (571)272-7433. The examiner can normally be reached M-F 7:30-4:30.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Cesar Paula can be reached at (571) 272-4128. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/D.T./Examiner, Art Unit 2145
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
Prosecution Insights