Image evaluation

Abstract

A machine can be configured to perform an image evaluation of an image depicting an item for sale and provide suggestions for improving the image depicting the item to increase the sales volume of the item depicted in the image. For example, the machine accesses the results of user behavior analysis. The machine receives an image of an item from a user device. The machine performs an image evaluation of the received image based on the analysis of the received image and the results of user behavior analysis. Performing the image evaluation can include determining the likelihood of a user engaging in a desired user behavior related to the received image. Then, the machine generates an output based on the evaluation of the received image, the output referring to the received image and indicating the likelihood of the user engaging in the desired behavior.

Description

[0001] This application is a divisional application of the invention patent application No. 201580018485.X entitled "Image Evaluation" after the PCT international application PCT / US2015 / 024112 entered China on April 2, 2015.

[0002] Priority

[0003] This patent claims the priority of U.S. Provisional Application No. 61 / 975,608 filed on April 4, 2014 and U.S. Patent Application No. 14 / 319,224 filed on June 30, 2014, and incorporates both of them herein by reference in their entireties.

[0004] Copyright

[0005] A portion of the disclosure of this patent document contains copyrighted material. The copyright owner does not object to the reproduction of the patent document or patent disclosure as it appears in the files or records of the Patent and Trademark Office by anyone, but the copyright owner reserves all copyrights. The following notice applies to the software and data described below and forming part of this document: Copyright eBay, Inc. 2014. All rights reserved. Technical Field

[0006] The subject matter disclosed herein generally relates to data processing. Specifically, the present disclosure relates to systems and methods that facilitate image evaluation. Background Art

[0007] Images depicting items for sale can be used to visually convey information about the items to potential buyers. For example, an online clothing store can use images to illustrate one or more merchandise items available for purchase at the online clothing store. The images can show clothing items made by people being worn on a model or a flat pattern being shown (e.g., neither a human model nor a mannequin). Brief Description of the Drawings

[0008] Some embodiments are illustrated in the drawings by way of example and not limitation.

[0009] Figure 1 is a network diagram depicting a client - server system in which some example embodiments can be deployed.

[0010] Figure 2 is a block diagram showing a marketplace and payment application provided as part of an application server 118 in a networked system in some example embodiments.

[0011] Figure 3 is a network diagram showing a network environment suitable for image evaluation according to some example embodiments.

[0012] Figure 4 It is a functional diagram of an example image evaluation machine according to some example embodiments.

[0013] Figure 5 It is a block diagram showing components of an image evaluation machine according to some example embodiments.

[0014] Figures 6 - 14 It is a flowchart showing operations of an image evaluation machine in a method of evaluating one or more images according to some example embodiments.

[0015] Figure 15 It is a block diagram showing a mobile device according to some example embodiments.

[0016] Figure 16 It is a block diagram showing components of a machine that can read instructions from a machine-readable medium and execute any one or more of the methods discussed herein. Detailed Description

[0017] There are provided methods and systems for evaluating images describing online sales of items to improve item sales volume. Examples merely represent possible variations. Unless explicitly stated otherwise, components and functions are optional and can be combined or subdivided, and operations can vary in sequence or be combined or subdivided. In the following description, for purposes of explanation, numerous specific details are set forth to provide a thorough understanding of example embodiments. However, it will be apparent to those skilled in the art that the subject matter may be practiced without these specific details.

[0018] Fashion (such as clothing) is a fast-growing type in online stores. Since users shopping in online stores may not have the same perceptual experience as when shopping in physical stores, the use of images describing items for sale (such as goods or services) is very important. Images of clothing items, especially their appearance, which are important to buyers, play a key role in conveying key information about the goods that is difficult to express in text. Whether a user selects (such as clicks on) an image to visually inspect the item shown in the image may affect whether the user purchases the item.

[0019] Since the visual presentation of items for sale (e.g., clothing items) can significantly influence a user's choice of a particular item (e.g., initial search, deciding to examine an image depicting the item, deciding to flag the image for future reference, or purchase decision), it may be beneficial for a seller to obtain an assessment of the relative effectiveness of different types of images when presenting items for sale. An assessment of images depicting items for online sale can facilitate presenting the item in the image in the best possible way such that the image serves as a persuasive tool to help the seller of the item sell the item. Additionally, an assessment of images depicting items for sale can assist the seller in avoiding potential misconclusions by some potential buyers who may consider the item of poor quality simply based on the quality of the item image at the price difference.

[0020] In some example embodiments, an image evaluation system can facilitate an image evaluation of an image depicting an item for sale. In some example embodiments, the image evaluation of the image can be based on image-based image analysis and an analysis of user behavior associated with (e.g., involved with) the image. According to certain example embodiments, the evaluation of the image can include examining the image based on multiple image attributes (e.g., display type of the item, professional photography, lighting, ambiance, image quality, and suitable combinations thereof), classifying the image into one or more categories based on one or more image attributes, comparing the image with other images submitted by other sellers, and / or determining the likelihood of obtaining a desired response from a user (e.g., a potential buyer of the item) viewing the image. Such an evaluation can be based on a comparison of the image with other similar items in the same category, as well as a comparison with images provided by other sellers.

[0021] The image evaluation can be used to provide feedback or suggestions to the provider of the image (e.g., the seller of the item for sale) with respect to the image in order to increase the sales volume of the item described in the image (e.g., improve the sales rate). The feedback can include statements such as: "Your item image is 80% better than the images uploaded by other sellers of the item"; "Your item image is 40% better than the images provided by sellers of this type of item"; "Your item image will be improved if you use a model to display the item described in the image"; "Use better lighting"; "Use higher contrast in the image"; or "Use a different color filter".

[0022] According to various embodiments, recommendations can be made to improve an image in order to obtain a desired response from a user (e.g., increase the click-through rate of the image, or increase the sales volume of an item). In some example embodiments, the image evaluation of an image submitted (e.g., received from) by a seller of an item depicted in the image is based on the analysis result of the image submitted by the seller and the data analysis result of user behavior related to images depicting items similar to the item depicted in the image received from the seller (e.g., clothing categories with the same terms). The data describing user behavior related to the image can include a representation of the actions taken by a user (e.g., a buyer) in response to seeing a plurality of images presented to the user. The data describing user behavior related to the image (also referred to as "user behavior data" or "user behavior representation") can be based on the interaction of one or more users with a multitude of images including various display types and varying image qualities (e.g., lighting, professional photography, and their suitable combinations).

[0023] In some example embodiments, large-scale user behavior data from e-commerce platforms around the world can be used to evaluate the effectiveness of different display types on users' shopping behavior. Generally, in the scenario of online clothing sales, clothes can be displayed in three ways: on a human model, on a mannequin, or flat display (e.g., without a mannequin or a human model). Analysis of behavioral data and transaction data (e.g., clicks, views, tags, or purchases) reveals that users are more attracted to clothes modeled on a human model than those displayed on a mannequin or in a flat mode, even when considering other factors (e.g., price or buyer details). In some example embodiments, the image evaluation system predicts the attention level of a user towards an image depicting an item (e.g., a clothing item) by modeling user preferences. The image evaluation system can also determine the likelihood of purchasing the item based on the item display (e.g., one or more images depicting the item) provided by the seller of the item. In some cases, the image evaluation system can recommend to the buyer a display type that is more effective in attracting the buyer's attention to the item and can increase the sales rate of the item.

[0024] Figure 1 is a network diagram depicting a client-server system 100 in which an example embodiment can be deployed. A networked system 102 in the form of an example of a web-based marketplace or publishing system provides server-side functionality to one or more clients via a network 104 (e.g., the Internet or a wide area network (WAN)). For example, Figure 1 shows a web client 106 (e.g., a browser, such as the Internet Explorer browser developed by Microsoft Corporation of Redmond, Washington) and a programmed client 108 executing on various devices 110 and 112.

[0025] The Application Programming Interface (API) server 114 and the network server 116 are connected to one or more application servers 118 and provide programming and network interfaces to the one or more application servers respectively. The application servers 118 host one or more marketplace applications 120 and payment applications 122. The application servers 118 are shown as being further coupled to one or more database servers 124 that facilitate access to one or more databases 126.

[0026] The marketplace application 120 can provide multiple marketplace functions and services to users accessing the networked system 102. In various example embodiments, the marketplace application 120 can include an image evaluator 132. In some example embodiments, the image evaluator 132 can facilitate the evaluation of images depicting items for sale and determine the likelihood of obtaining a desired response from users (e.g., potential buyers of the item) viewing the images.

[0027] The payment application 122 can similarly provide multiple payment services and functions to users. The payment application 122 can allow users to accumulate value in an account (e.g., circulating currency (such as US dollars) or proprietary currency (such as "points")), and subsequently redeem the accumulated value for products (e.g., goods or services) that can be purchased through the marketplace application 120. Although Figure 1 both the marketplace application 120 and the payment application 122 are shown as forming part of the networked system 102, it will be understood that in alternative embodiments, the payment application 122 can form part of a payment service separate and remote from the networked system 102.

[0028] Furthermore, although Figure 1 the system 100 shown in

[0029] employs a client-server architecture, of course, the embodiments are not limited to such an architecture, but can equally well be applied to, for example, distributed or peer-to-peer architecture systems. The various marketplace applications 120 and payment applications 122 can also be implemented as stand-alone software programs that do not necessarily have networking capabilities.

[0030] Figure 1 Also shown is a third-party application 128 executing on a third-party server machine 130, which third-party application 1043 is capable of programmatically accessing the networked system 102 via a programming interface provided by the API server 114. For example, the third-party application 128 can utilize information obtained from the networked system 102 to support one or more features or functions on a third-party-hosted website. For example, the third-party website can provide one or more promotional, marketing, or payment functions supported by a relevant application of the networked system 102.

[0031] Figure 2 FIG. is a block diagram showing a marketplace application 120 and a payment application 122 provided as part of an application server 118 in a networked system 102 in an example embodiment. The applications 120 and 122 can reside on dedicated or shared server machines (not shown) that are communicatively coupled to enable communication between the server machines. The applications 120 and 122 are communicatively coupled to each other (e.g., via appropriate interfaces) and to various data sources, thereby allowing information to be passed between the applications 120 and 122 or thereby allowing the applications 120 and 122 to share and access common data. The applications 120 and 122 can also access one or more databases 126 via a database server 124.

[0032] The networked system 102 can provide numerous listing, posting, and price-setting mechanisms, where sellers can list goods or services for sale (or post information about goods or services), buyers can express interest or intent to purchase such goods or services, and a price can be set for a transaction involving the goods or services. To this end, the marketplace application 120 and the payment application 122 are shown as including at least one posting application 200 and one or more auction applications 202, which support auction-format listings and price-setting mechanisms (such as English auctions, Dutch auctions, second-price auctions, Chinese auctions, combinatorial auctions, reverse auctions, etc.). The various auction applications 202 can also provide multiple functions to support such auction-format listings, such as a reserve price function (whereby a seller can specify a reserve price associated with a listing) and a proxy bid function (whereby a bidder can invoke an automated proxy bid).

[0033] A plurality of fixed-price applications 204 support fixed-price listing formats (such as traditional classified-ad-style listings or catalog listings) and buyout listings. Specifically, buyout listings can be provided in conjunction with auction-format listings (such as including the now buy (BIN) technology developed by eBay Inc. of San Jose, Calif.), and allow users to purchase goods or services also being sold via auction at a fixed price that is typically higher than the starting price of the auction.

[0034] The Store Application 206 allows a seller to group listings in a "virtual" store, which can be branded or personalized by the seller. Such a virtual store can also offer promotions, incentives, and specialized features personalized by the relevant seller.

[0035] The Reputation Application 208 allows trading users to establish, build, and maintain a reputation using the networked system 102, which can be published and made available to potential trading partners. Given situations such as the networked system 102 enabling person-to-person trading, otherwise users may not have a history or other reference information to evaluate the credibility and creditworthiness of potential trading partners. The Reputation Application 208 allows users (e.g., through feedback provided by other trading partners) to build a reputation over time in the networked system 102. Other potential trading partners can then refer to this reputation to assess credit and credibility.

[0036] The Personalization Application 210 allows users of the networked system 102 to personalize various aspects of their interaction with the networked system 102. For example, a user can create a personalized reference page using the appropriate Personalization Application 210, where information about transactions the user is (or was) a party to can be viewed. Additionally, the Personalization Application 210 can enable users to personalize listings and other aspects of their interaction with the networked system 102 and other parties.

[0037] The networked system 102 can support multiple markets customized for specific geographic regions, for example. One version of the networked system 102 can be customized for the United Kingdom, while another version of the networked system 102 can be customized for the United States. Each of these versions can operate as a separate market or can be a customized (or internationalized) presentation of a common underlying market. The networked system 102 can thus include multiple Internationalization Applications 212 that customize information (and / or the presentation of information by the networked system 102) according to predetermined criteria (such as geographic, demographic, or market criteria). For example, the Internationalization Applications 212 can be used to support the customization of information for multiple regional websites operated by the networked system 102 and accessible via the respective web servers 116.

[0038] Navigation of the networked system 102 can be implemented by one or more Navigation Applications 214. For example, a search application (as an example of a Navigation Application 214) can implement keyword searches of listings published via the networked system 102. A browsing application can allow a user to browse through various categories, directories, or inventory data structures according to which listings can be classified in the networked system 102. Various other Navigation Applications 214 can be provided to complement the search and browsing applications.

[0039] To make the lists available via the networking system 102 as intuitive and eye-catching as possible, the applications 120 and 122 can include one or more imaging applications 216 that a user can utilize to upload images to include in the lists. The imaging application 216 also operates to incorporate the images in the viewing lists. The imaging application 216 can also support one or more promotional features, such as a gallery presented to potential buyers. For example, a seller can pay an additional fee to include an image in the gallery of a promoted item.

[0040] The listing creation application 218 allows buyers to conveniently create lists related to the goods or services they want to transact via the networking system 102, and the listing management application 220 allows sellers to manage these lists. Specifically, managing these lists can be a challenge in cases where a particular seller has created and / or published a large number of lists. The listing management application 220 provides multiple functions (such as automatic relisting, inventory level monitoring, etc.) to assist the seller in managing these lists. One or more post-listing management applications 222 also help the seller with multiple activities that typically occur after a listing. For example, when an auction assisted by one or more auction applications 202 is completed, the seller may wish to leave feedback about a particular buyer. To this end, the post-listing management application 222 can provide an interface to one or more reputation applications 208, allowing the seller to conveniently provide feedback about multiple buyers to the reputation application 208.

[0041] The dispute resolution application 224 provides a mechanism that can resolve disputes arising between the trading parties. For example, the dispute resolution application 224 can provide a guided process that leads the trading parties through multiple steps to attempt to settle the dispute. If the dispute cannot be settled via the guided process, the dispute can be escalated to a third-party mediator or arbitrator.

[0042] Multiple anti-fraud applications 226 implement fraud detection and prevention mechanisms to reduce fraud occurring in the networking system 102.

[0043] The messaging application 228 is responsible for generating messages and delivering them to the users of the networking system 102 (e.g., notifying users of messages related to the status of lists on the networking system 102 (such as providing a "you've been outbid" notification to bidders during an auction, or providing promotional and marketing information to users)). Each messaging application 228 can utilize any of multiple messaging networks and platforms to deliver messages to users. For example, the messaging application 228 can deliver e-mail, instant message (IM), short message service (SMS), text, fax, or voice (such as voice over IP (VoIP)) messages via a wired network (such as the Internet), plain old telephone service (POTS), or a wireless (such as mobile, cellular, WiFi, WiMAX) network 104.

[0044] The merchandising application 230 supports various merchandising functions available to sellers so that the sellers can increase their sales volume through the networked system 102. The merchandising application 230 also operates various merchandising functions that can be invoked by the sellers and can monitor and track the achievements of the merchandising strategies employed by the sellers.

[0045] The networked system 102 itself, or one or more parties transacting through the networked system 102, can operate a loyalty program supported by one or more loyalty / promotion applications 232. For example, a buyer can earn loyalty or promotion points for each transaction established and / or completed with a specific seller and can obtain rewards by redeeming the accumulated loyalty points.

[0046] Figure 3 is a network diagram showing a network environment 300 suitable for image evaluation according to some example embodiments. The network environment 300 includes an image evaluation machine 310 (e.g., image evaluator 132), databases 126 and 330 and 350, all of which are communicatively coupled to each other via a network 390. The image evaluation machine 310, with or without the database 126, can form part or all of a network-based system 305 (e.g., a cloud server-based system configured to provide one or more image processing services, image evaluation services, or both services to devices 330 and 350). One or both of the devices 330 and 350 can include a camera that allows images (e.g., images of items for sale) to be taken. One or both of the devices 330 and 350 can facilitate the transmission of the images (e.g., submitted to the database 126) to the image evaluation machine 310. As described below with reference to Figure 16 described, the image evaluation machine 310, devices 330 and 350 can be implemented in a computer system, either wholly or in part, respectively.

[0047] Also in Figure 3Users 332 and 352 are shown. One or both of users 332 and 352 can be a human user (e.g., a human), a machine user (e.g., a computer configured by a software program to interact with device 330), or any suitable combination thereof (e.g., a machine-assisted human or a human-supervised machine). User 332 is not part of network environment 300, but is associated with device 330 and can be a user of device 330. For example, device 330 can be a desktop computer, an in-vehicle computer, a tablet computer, a navigation device, a portable media device, a smart phone, or a wearable device (e.g., a smart watch or smart glasses) belonging to user 332. Similarly, user 352 is not part of network environment 300, but is associated with device 350. As an example, device 350 can be a desktop computer, an in-vehicle computer, a tablet computer, a navigation device, a portable media device, a smart phone, or a wearable device (e.g., a smart watch or smart glasses) belonging to user 352.

[0048] Figure 3 Any machine, database, or device shown therein can be implemented by a general-purpose computer modified by software (e.g., one or more software modules) to be a special-purpose computer to perform the functions described herein for that machine, database, or device. For example, a computer system capable of implementing any one or more of the methods described herein is discussed below with reference to Figure 16 As used herein, a "database" is a data storage resource and can store data structured as text files, tables, spreadsheets, relational databases (e.g., object-relational databases), triple stores, integrated data stores, or any suitable combination thereof. Additionally, Figure 3 any two or more of the machines, databases, or devices shown therein can be combined into a single machine, and the functions described herein for any single machine, database, or device can be re-divided among multiple machines, databases, or devices.

[0049] Network 390 can be any network that supports communication between machines, databases, and devices (e.g., server 310 and device 330). Thus, network 390 can be a wired network, a wireless network (e.g., a mobile or cellular network), or any suitable combination thereof. Network 390 can include one or more portions that build a private network, a public network (e.g., the Internet), or any suitable combination thereof. Thus, network 390 can include a combination of a local area network (LAN), a wide area network (WAN), the Internet, a mobile telephone network (e.g., a cellular network), a wired telephone network (e.g., a plain old telephone system (POTS) network), a wireless data network (e.g., a WiFi network or a WiMax network), or any suitable combination. Any one or more of network 390 can communicate information via a transmission medium. As used herein, "transmission medium" refers to any non - tangible (e.g., transient) instruction that can transmit (e.g., emit) instructions executed by a machine (e.g., by one or more processors of such a machine), and includes digital or analog communication signals or other transient media to facilitate the transmission of such software.

[0050] Figure 4 is a functional diagram of an example image evaluation machine 310 according to some example embodiments. In some example embodiments, the image evaluation machine 310 is included in a network - based system 400. As described in more detail below, the image evaluation machine 310 can receive an image 410 (e.g., an image of a clothing item). The received image 410 (also referred to as "image 410") can be received from a device 330 associated with a user 332.

[0051] In response to receiving the image 410, the image evaluation machine 310 uses at least one computer processor to analyze the image (e.g., perform image analysis 420 of the image). In some example embodiments, to perform the image analysis 420, the image evaluation machine 310 extracts one or more visual features from the received image 410. The image evaluation machine 310 can also identify values of one or more image attributes of the received image 410 based on the one or more features of the received image 410 that are extracted and the image attribute data 430 (e.g., data that identifies or describes the image attributes and the values that the image attributes can take). The image evaluation machine 310 can also classify the received image 410 into one or more categories based on the values of the one or more image attributes 430.

[0052] For example, one or more visual features extracted from the received image 410 may include data representing a display type (e.g., human, mannequin, or flat display) for displaying a clothing item depicted in the received image 410. The image attribute data 430 may include a variety of image attributes such as display type, background, contrast, or illumination. Each image attribute 430 may be associated with one or more values (e.g., acquire or have corresponding values). For example, the display type attribute may take one of three values: (1) human, (2) mannequin, or (3) flat display. Based on one or more visual features extracted from the received image 410 and the display type attribute, the image attribute evaluator 310 may identify the value of the display attribute of the received image 410 (e.g., human). The image evaluator 310 may classify the received image 410 into a specific display type category (e.g., human category, mannequin category, or flat display category) based on the identified value of the display type attribute of the received image 410.

[0053] In some example embodiments, instead of or in addition to classifying the received image 410 into a category based on the value of the display type attribute of the received image 410, the image evaluator 310 assigns (e.g., attributes to) a label (e.g., tag) to the received image 410 that identifies the display type for displaying the item depicted in the image. The image evaluator 310 may also calculate a confidence score value for the received image 410 to represent the level of confidence in the correct determination of the display type of the received image 410. The label or the confidence score value or both may be stored in a record of a database (e.g., database 126) in association with the received image 410 and may be used for future image evaluation of the received image 410.

[0054] According to certain example embodiments, one or more image attributes and their values (e.g., one or more attribute-value pairs) are used as a basis for calculating an image score value of an image. For example, for a specific image of a black dress, the value of the attribute "display type" is identified as "human", the value of the attribute "illumination" is determined as "good", and the value of the attribute "background" is determined as "white". Based on these pairs of attribute values, an image score value may be determined for the specific image of the black dress. Other attributes such as contrast, sharpness, arrangement, composition, or balance or a suitable combination of these attributes may also be used as a basis for calculating an image score value of an image depicting an item.

[0055] In some example embodiments, the image score value of the image may be based on the trust score value of the image and the low-level quality score value of the image. The image evaluator 310 may determine a trust score attributable to (e.g., for, of, or assigned to) the received image 410, the trust score measuring the confidence level of classifying the received image 410 into the class to which the received image 410 belongs (e.g., an image using the "human" display type belongs to the human class). In certain example embodiments, the trust score value of the received image 410 takes a value between 0 and 1. The higher the trust score value, the higher the confidence level that the image evaluator 310 correctly determines the classification of the received image 410 into the class. For example, the image evaluator 310 determines that image A is of the "mannequin" display type with a trust score value of.8. This may mean that with 80% confidence, the image evaluator 310 correctly determines the display type of image A. The image evaluator 310 may determine a low-level quality score value attributable to (e.g., for, of, or assigned to) the received image 410 based on one or more other image attributes of the received image 410 (e.g., lighting, clarity, or professional production). In some example embodiments, the trust score value of the received image 410 and the low-level quality score value of the received image 410 are combined to produce the image score value of the received image 410.

[0056] In some examples, the trust score value of the received image 410 and the low-level quality score value of the received image 410 may be combined by multiplying the trust score value of the received image 410 and the low-level quality score value of the received image 410 to calculate the image score value of the received image 410. In other examples, the trust score value of the received image 410 and the low-level quality score value of the received image 410 may be specifically weighted according to a weight assignment rule to produce a weighted trust score value of the received image 410 and a weighted low-level quality score value of the received image 410. The weighted trust score value of the received image 410 and the weighted low-level quality score value of the received image 410 may be added together to calculate the image score value of the received image 410. In some example embodiments, specific weighting may be selected in the user behavior analysis 450 device of the user behavior data 440, the user behavior data representing how a user who sees multiple images depicting similar items and having different image attribute values acts in relation to a specific image.

[0057] In some example embodiments, the image evaluation machine 310 grades images included in a specific classification of an image (e.g., an image depicting clothing in the human category) or a sub - category of a specific category of an image (e.g., an image depicting a black dress in the task category). The grading can be based on a trust score value of the image, a low - level quality score value of the image, or an image score value that combines the trust score value and the low - level quality score value of the corresponding image. For example, the image evaluation machine 310 identifies an image included in the human category that depicts a clothing item shown on a human. The image evaluation machine 310 determines the trust score value and the low - level quality score value of the corresponding image in the human category. For each image in the human category, the image evaluation machine 310 can combine the trust score value and the low - level quality score value corresponding to the specific image to generate an image score value corresponding to the specific image. When calculating the image score value for an image in the human category, the image evaluation machine 310 can grade the images within the human category based on their respective image score values. In some example embodiments, the images are presented to a user (e.g., a buyer) according to their image score values.

[0058] According to various example embodiments, the image evaluation machine 310 collects (e.g., takes, obtains, or receives) user behavior data 440 (e.g., an indication of actions taken by potential buyers, actual buyers, or a combination thereof related to multiple images depicting a specific type of clothing item). In some example embodiments, the user behavior data 440 represents the behavior of one or more users searching for an item and reacting (e.g., selecting or not selecting, viewing or not viewing) to one or more images depicting the searched - for item and shown to the one or more users. In some example embodiments, the user behavior data 440 is collected over a period of time. In other example embodiments, the user behavior data 440 is collected at a particular moment.

[0059] The image evaluation machine 310 can perform user behavior analysis 450 based on user behavior data 440 to learn (e.g., determine) what images the user (e.g., the buyer) prefers. Specifically, the image evaluation machine 310 can determine which image attribute-value pairs may be correlated with the desired user behavior (e.g., clicking on an image or purchasing an item described in the image). For example, the user behavior data 440 can be collected from an e-commerce platform (e.g., an online marketplace, an online store, or a website). The e-commerce platform can enable users to search for items for sale using text queries entered on the website. In each search session, the user can enter a query to find an item, and the search engine can return multiple search results (e.g., images depicting the item) that mention the item. The search can be a very personalized task with great variation in terms of search purpose and product attributes. To focus only on clothing items and limit the images to similar content (e.g., products of the same category), the image evaluation machine 310 can display the range of the collected user behavior data.

[0060] In some example embodiments, the image evaluation machine 310 can be configured to be a query-dependent machine. As a query-dependent machine, the image evaluation machine 310 can be configured to collect only the descriptions of query conversations that use a specific keyword (e.g., data describing the query conversations). For example, the image evaluation machine 310 can be configured to collect only the descriptions of query conversations that use the keyword "black dress". In another example, the image evaluation machine 310 can be configured to collect only the descriptions of query conversations that use the keyword "black dress". By configuring the image evaluation machine 310 to be a query-dependent machine, the image evaluation machine 310 can display the range of user behavior data collected based on a specific query.

[0061] The image evaluation machine 310 can rank the retrieval results (e.g., images depicting black dresses) of the items based on the relevance relationship of the items to a specific keyword. In some example embodiments, when collecting the user behavior data 440, the image evaluation machine 310 can collect only the data of user behavior that describes items with high relevance. High-relevance items can be the items described in the images displayed on the first retrieval result page in response to the query entered by the user.

[0062] Therefore, the images that are part of the retrieval results are likely to have the same content (e.g., black dress). The images may only differ in their presentation (e.g., image attributes). For example, the images may differ in the display type of the black dress: on a human model, a mannequin, or just a flat display. It is beneficial for the seller to know how user behavior is related to images with similar content but different image attributes.

[0063] The execution of user behavior analysis 450 can indicate how user behavior is manifested when presented with multiple images having different image attributes. In some examples, the execution of user behavior analysis 450 can facilitate determining whether an image is pleasing to a buyer, whether the image is likely to help sell the item shown in the image, or whether the image is likely to be overlooked by potential buyers of the item described in the image. In some example embodiments, the execution of user behavior analysis 450 can include: identifying specific user actions of one or more users in relation to one or more images that depict an item of clothing (such as a black dress) and manifest one or more user interests in the image or the item described in the image. In some examples, user interest in a specific image or the item described in a specific image can be inferred based on the user's interaction with the specific image. The specific user actions can include, for example, selecting (such as clicking) a specific image from a plurality of images presented to the user, marking the specific image for later reference, emailing the image, printing the image, viewing the image for a time period exceeding a threshold, or purchasing the item described in the image within another threshold time period after viewing the image. The execution of user behavior analysis 450 can also include identifying one or more attributes of one or more images that depict items of clothing, and determining whether the one or more attributes associate increased user interest - manifested activity with the one or more images.

[0064] In some example embodiments, user behavior analysis 450 can focus on answering the following questions with respect to display type attributes: (1) Do the three styles (e.g., human, mannequin, or flat display) have different likelihoods of being clicked on a search results page? (2) Do the three styles differ in stimulating users to mark or "view" an item? (3) Do the three styles differ in increasing the sales rate? (4) How much difference in quantity exists between the preferences of users for the three styles? Other or additional questions can be used when examining the impact of other image attributes on user purchasing behavior.

[0065] For a successful transaction, attracting the attention of potential consumers can be crucial. User interest can be shown at different stages during the online shopping process (e.g., browsing, click actions, or purchases). A user selection model (also referred to as the "PMF - user selection model" or "PMF model") for quantifying user preferences among humans, mannequins, and flat display styles can facilitate the understanding and quantification of user responses at three different stages during the online shopping cycle: (a) the "click" action at the search results page, where multiple relevant items can be displayed according to a search query; (b) the "view" action at the view item page, where the buyer can evaluate the item in more detail and can make decisions to hold the item (e.g., by viewing), continue browsing, or purchase the item; or (c) the "purchase" action, where the user makes a final decision on the product.

[0066] Type Displayed item Clicked item Unclicked item Flat display 40.87％ 39.21％ 40.99％ Mannequin 34.49％ 33.26％ 34.57％ Human 24.65％ 27.53％ 24.44％

[0067] Table 1. Distribution shift for displayed, clicked, and non - clicked items. For the clicked items, the proportion of the P type increases while the proportions of the M type and F type decrease, indicating that users prefer the P type over the M type or F type.

[0068] Given the multiple relevant items displayed on the search results page, the user click response at the search results page can be identified (e.g., by the image evaluator 310). By classifying the image content into PMF types, Table 1 above shows a significant distribution shift from the originally displayed search results to the content clicked by the user. The proportion of the human type (also referred to as the "P type") is only 24.65% for the displayed items but increases to 27.53% for the clicked items. For the clicked items, the proportions of both the mannequin type (also referred to as the "M type") and the flat type (also referred to as the "F type") decrease. This distribution indicates that users prefer the P type display over the M type or F type. Even for different price segments or seller types, buyers show a stronger preference for items presented in the P type.

[0069] Assume that on the search result page, the P type attracts more attention. The image evaluation machine 310 can browse the user actions on the item page. Browsing the item page can be a page where the user can obtain details about the item described in the image and can participate in other interest-revealing actions that indicate a more direct shopping intention (e.g., marking the item for more careful evaluation). The image evaluation machine 310 can calculate the average number of views for each PMF type and for each seller group. The results shown in Table 2 below suggest a positive correlation between the "viewing action" and the top sellers and the P type product display. For items sold by temporary sellers or top sellers, the P type image helps increase the chance of viewing. When compared with less-viewed items, the proportion of P type images increases for items with a higher viewing frequency.

[0070]

[0071] Table 2. Average "number of views" for each display type relative to seller type. The results suggest a correlation between the P type and a higher average viewing rate for both temporary sellers and top sellers.

[0072] The sales rate can be a limiting evaluation metric for prior listings. Table 3 lists the conversion rates for each display type grouped by click actions observed in the collected session data. Compared with unclicked items, clicked items show a higher conversion rate because the user shows interest in the item by clicking on the image showing the item, which is expected and leads to a higher chance of purchase. A comparison of the three display types (e.g., human, mannequin, and flat display) can show that items displayed in the P type demonstrate a better sales rate for both clicked and unclicked items.

[0073]

[0074] Table 3. Conversion rates for three display types for clicked and unclicked items in the collected session data, where items displayed in the P type show a better sales rate.

[0075] The image evaluation machine 310 can also use the PMF-user selection model to determine and quantitatively compare the differences in preferences for each display style. In the PMF-user selection model, m, p can represent the preference level for each type, m, p can be the proportion of each type in the original set of retrieved items, where F f +F m +F p = 1, and m and p can be the proportions of each type in the clicked item. The smaller the proportion Fi, the more difficult it is to be displayed in the retrieval results and selected by the user. The preference W i is higher, the more likely this given type will be selected. Therefore, P i is affected by two factors: the distribution bias represented by F i and the preference bias W i .

[0076] There may be different ways to combine these two factors to obtain a reasonable prediction of the user preference level. In some example embodiments, only two choices are weighted, and it can be assumed that the difficulty in selecting a given type is inversely proportional to the proportion of their post-click distribution, where the difficulty is modeled by both Fi and W i . It can also be assumed that the preferences are equal (e.g., W f = W m = W p ), and there should be no significant shift between the pre-click distribution and the post-click distribution. In other words, it can be expected that F i = P i . Based on this idea, the PMF user selection model one (C1) can be proposed as;

[0077]

[0078] where when the levels of all three types are the same, results in the same PMF distribution for both pre-click and post-click data. Given the same constraints, another model can instead use multiplication, resulting in model two (C2):

[0079]

[0080] The solutions of these two models are a set of paired relationships between the preference levels, where the W f and W m parameters are parameterized as functions of W p . By adopting the inputs in Table 1, i.e., assigning F f = 0.4087, F m = 0.3449, and F p = 0.2465 as the pre-click distribution, and assigning P f = 0.3921, P m = 0.3326, and P p = 0.2753 as the post-click distribution, model C1 produces:

[0081]

[0082]

[0083] The result of Model C2 directly shows that the preference for M type is approximately 86.3% of the preference for P type:

[0084]

[0085]

[0086] Table 4 lists the prediction priority levels for two models according to the exemplary embodiments and the sum of which is one. The P type has the highest priority, and no significant difference is found between the M type and the F type. There are two possible reasons. First, the flat category includes many non-clothing items retrieved when a user uses a query like "black dress shoes". E-shoppers may tend to click on those items because they explore or search for similar items that match well with the black dress (e.g., shoes or belts). Second, since the mannequin is an inanimate humanoid figure, it cannot be comfortably regarded as an actual human by viewers.

[0087] Type C1 C2 Flat display 0.3147 0.3155 Mannequin 0.3133 0.3171 Human 0.3740 0.3673

[0088] Table 4. Preference levels for each PFM type estimated by two proposed PMF user selection models, where the M type or the F type is approximately 86% of the P type.

[0089] The results of the user behavior analysis 450 as described above can show that in some exemplary embodiments, the P type display of the clothes in the image is the most effective product display among the three display types, and can maximize helping to attract the user's attention and increase the sales rate. The results of the user behavior analysis 450 can be used for multiple audiences, for example, for clothing e-retailers to select better display strategies, or for e-commerce operators to design higher retrieval or supply recommendation systems to improve the click-through rate. Additionally, in some exemplary embodiments, the user behavior analysis 450 can be used to evaluate an image depicting an item to determine the likelihood that the item elicits the desired response from potential buyers of the item.

[0090] For example, user behavior analysis 450 and image analysis 420 can be used to perform an image evaluation 460 of the received image 410 received from the device 130 of the user 132 (e.g., a seller). In some example embodiments, the image evaluation machine 310 can calculate an average likelihood that a buyer will engage in a performance action of interest related to (e.g., involving) the received image 410 or the item described in the received image 410 based on the results of the image analysis 420 and the results of the user behavior analysis 450. Examples of performance actions of interest related to the received image 410 or the item described in the received image 410 are selecting (e.g., clicking) the received image 410 depicting the item or purchasing the item while viewing the received image 410. According to certain example embodiments, the image evaluation machine 310 determines an image score value for the received image 410, which can represent to the user 132 how an average buyer will respond to the received image 410 (e.g., what is the likelihood that a buyer will click on the received image 410 or what is the likelihood that a buyer will purchase the item shown in the image).

[0091] According to some example embodiments, when performing the image evaluation 460, the image evaluation machine 310 can generate an output 470 that can be transmitted to the user 132 via the device 130. Examples of the output 470 include the results of the image evaluation 460 fed back based on the results of the image evaluation 460, or suggestions on how to improve the display (e.g., showing) of the item included in the received image 410. In some examples, the output 470 can include an image score value for the received image 410, a ranking of the received image 410 compared to other images provided by other sellers (e.g., within the category of the image or regardless of the category of the image), or a representation of the likelihood of selling the item shown in the received image 410 if the received image 410 is used as a display on an item e-commerce website.

[0092] In some example embodiments, output 470 includes a recommendation on how to select a cost-effective display type based on the results of image analysis 420, image evaluation 460, or both. For example, output 470 may include an image score value for the received image 410 or a ranking of the received image 410 compared to other images provided by other sellers (e.g., within the category of the image or independent of the category of the image), and may provide one or more options to improve the display of the item using the image describing the item based on the image score value, ranking value (e.g., position, order, or score), or both. In some examples, one of the options may be to select a different type of display (e.g., select type M instead of type F, or select type P instead of type M), where the change in the display type is cost-effective. In other examples where the cost of changing the display type is high, the recommended selection may be to improve other attributes of the image (e.g., lighting, professional photography, or a simple or white background).

[0093] Figure 5 is a block diagram showing components of an image evaluation machine 310 according to some example embodiments. The image evaluation machine 310 is shown as including a receiver module 510, an image analysis module 520, a behavior analysis module 530, an output module 540, and a communication module 550, all configured to communicate with each other (e.g., via a bus, shared memory, or switch).

[0094] Any one or more of the modules described herein may be implemented using hardware (e.g., one or more processors of a machine) or a combination of hardware and software. For example, any module described herein may configure a processor (a processor among one or more processors of a machine) to perform the operations described herein for that module. Additionally, any two or more of these modules may be combined into a single module, and the functions described herein for a single module may be subdivided among multiple modules. Further, according to various example embodiments, modules described herein as being implemented in a single machine, database, or device may be distributed among multiple machines, databases, or devices.

[0095] Figures 6 - 14 is a flowchart showing the operations of the image evaluation machine 310 in performing a method 600 for evaluating one or more images according to some example embodiments. The operations in method 600 may be performed using the modules Figure 5 described above. As Figure 6 shown, method 600 may include one or more of operations 610, 620, 630, and 640.

[0096] The image evaluation performed by the image evaluation machine 310 can begin with method operation 610, where the receiver module 510 accesses one or more results of user behavior analysis. The results of user behavior analysis can be generated by the behavior analysis module 530 based on the analysis of user behavior data 440. The user behavior data 440 can relate to multiple test images. The user behavior data 440 can be collected based on the user behavior of one or more users (e.g., potential buyers or actual buyers) associated with the multiple test images. The analysis of the user behavior data 440 can include determining the user preferences of one or more users for specific values of one or more image attributes for the multiple images (e.g., the images included in the test image library) shown to the multiple users.

[0097] In method operation 620, the receiver module 510 receives an image of an item. The image can be received from a user device (e.g., a smartphone) of a user (e.g., a seller). In some example embodiments, the image can depict a clothing item.

[0098] In method operation 630, the image analysis module 520 uses one or more processors to perform image analysis of the image. Performing image evaluation of the image can include performing image analysis of the image and evaluating the image based on the image analysis of the image. The evaluation of the image can be based on the results of the image analysis and the results of the user behavior analysis accessed by the receiver module 510. In some example embodiments, the evaluation of the image includes determining the likelihood of obtaining a desired response from one or more buyers to whom the image is shown.

[0099] In method operation 640, the output module 540 generates an output that references (e.g., includes a citation, uses an identifier (ID) to identify, etc.) the image. The output can be generated based on the evaluation of the image. The output is generated for the user device in response to receiving the image of the item from the user device.

[0100] In some example embodiments, the method 600 can further include transmitting a communication (e.g., via the communication module 550) to the seller's device. The communication can include the output generated with reference to the image. More details regarding the method operations with respect to method 600 are described below. Figures 6A - 14 Description of more details regarding the method operations with respect to method 600.

[0101] As Figure 6AAs shown, according to some example embodiments, method 600 may include one or more of operations 631, 632, and 633. Method operation 631 may be performed as part of method operation 630 (e.g., a precursor task, a subtask, or a part), where output module 540 generates an output referencing the image. In method operation 631, image analysis module 520 calculates a score value for the received image. The calculation of the score value for the image may be based on the values of one or more image attributes of the image.

[0102] Method operation 632 may be performed after method operation 631. In method operation 632, image analysis module 520 determines the likelihood that a different user (e.g., a buyer) engages in a desired user behavior related to the received image. Examples of desired user behaviors related to the received image are selecting, clicking, or marking for future reference of the received image; purchasing an item described in the received image or placing the corresponding item on a wish list; and so on. In some examples, the received image may be received from a seller of an item described in the received image, from an agent of the seller, or from a user device of the seller or the seller's agent. In some example embodiments, determining the likelihood that a different user engages in a desired user behavior related to the received image may be based on the score value of the received image. In various example embodiments, determining the likelihood that a different user engages in a desired user behavior related to the received image may be based on one or more results of user behavior data analysis. In certain example embodiments, determining the likelihood that a different user engages in a desired user behavior related to the received image may be based on both the score value of the received image and one or more results of the user behavior data analysis.

[0103] Method operation 633 may be performed as part of method operation 640 (e.g., a precursor task, a subtask, or a part), where output module 540 generates an output referencing the image. In method operation 633, output module 540 may generate an output referencing the image and indicating the likelihood of user engagement in the desired user behavior related to the received image.

[0104] As Figure 6BAs shown, according to some example embodiments, method 600 may include one or more of operations 634, 634, 636, and 637. Method operation 634 may be performed after method operation 632, where the image analysis module 520 determines the likelihood that a user (e.g., a general buyer, a specific buyer, etc.) engages in the required user behavior related to the received image. In method operation 634, the image analysis module 520 accesses one or more image attribute values of another (e.g., second) image of another similar item. The item described in the image and the other similar items described in the other images may have similar characteristics, such as style, color, pattern, etc. For example, both the item and the other type of item may be little black dresses.

[0105] Method operation 635 may be performed after method operation 634. In method operation 635, the image analysis module 520 calculates a second score value corresponding to the other image. The calculation of the second score value may be based on the values of one or more image attributes of the other image.

[0106] Method operation 636 may be performed after method operation 635. In method operation 636, the image analysis module 520 compares the image with the other images based on the score value (e.g., the first score value) corresponding to the image and the second score value corresponding to the other images.

[0107] Method operation 637 may be performed after method operation 636. In method operation 637, the image analysis module 520 generates a ranking of the image related to the other images of other similar items.

[0108] As Figure 7 As shown, according to some example embodiments, method 600 may include one or more of operations 701, 702, 703, and 704. Method operation 701 may be performed before method operation 610, where the receiver module 510 accesses one or more results of user behavior analysis. In method operation 701, the behavior analysis module 530 accesses (e.g., receives) user behavior data. The user behavior data may be an indication of the actions taken by a user (e.g., an actual or potential buyer) in response to receiving descriptions of one or more similar items, an indication of no action taken, or a combination thereof.

[0109] Method operation 702 may be performed before method operation 610, where the receiver module 510 accesses one or more results of user behavior analysis. In method operation 702, refer Figure 2 As described above, the behavior analysis module 530 analyzes the user behavior data.

[0110] Method operation 703 may be performed before method operation 610, where receiver module 510 accesses one or more results of user behavior analysis. At method operation 703, behavior analysis module 530 generates one or more results of the analysis of the user behavior data. In some example embodiments, the one or more results of the analysis of the user data include one or more indications of user preferences for a specific image with respect to specific image attributes. For example, an indication of user preference represents that a user (e.g., a clothing buyer) generally prefers to use an image of a task display type to display clothing items. According to another example, an indication of user preference indicates that most users prefer an image with a white background using an image of a mannequin. User preference for a specific image may be implied based on the user's actions towards the specific image, such as selecting (e.g., clicking) a specific image or flagging an image for future reference. In some examples, user preference for a specific image may be implied based on the user purchasing an item described in the specific image.

[0111] Method operation 704 may be performed before method operation 610, where receiver module 510 accesses one or more results of user behavior analysis. At method operation 704, behavior analysis module 530 stores the analysis results of the user behavior data in a database (e.g., database 126).

[0112] As Figure 8 shown, according to some example embodiments, method 600 may include one or more of operations 801, 802, and 803. Method operation 801 may be performed as part of method operation 630 (e.g., a precursor task, a subtask, or a part), where image analysis module 520 performs an image evaluation of the image. At method operation 801, image analysis module 520 extracts one or more visual features from the image received from the seller.

[0113] Method operation 802 may be performed after method operation 801. At method operation 802, image analysis module 520 identifies the values of one or more image attributes of the image (e.g., the value of the display type for displaying an item within the image). Identifying the display type for displaying an item within the image may be based on one or more visual features extracted from the image received from the seller. In some example embodiments, method operations 801 and 802 are performed as part of image analysis 420 as described above. Figure 4 Method operations 801 and 802 are performed as part of image analysis 420 as described above.

[0114] Method operation 803 may be performed after method operation 802. In method operation 803, image analysis module 520 determines the likelihood that a user (e.g., a buyer) viewing the image participates in an interest display action (e.g., clicking on the image or purchasing an item shown in the image) related to the image. Determining the likelihood that a user viewing the image participates in an interest display action related to the image may be based on image analysis (or the results of image analysis, such as the values of one or more image attributes of the image, e.g., the value of the display type of the identified image) and user behavior data (e.g., the results of user behavior analysis). According to some example embodiments, the user behavior data includes data identifying the interactions of potential buyers of the item with one or more images showing the item. In various example embodiments, the buyer behavior data may represent the preferences of potential buyers when selecting images of a specific display type from among multiple display types. Examples of interest display actions interacting with the image are selecting or clicking on the image, browsing the image, placing the item shown in the image in a wish list, locking the image, or purchasing the item shown in the image.

[0115] As Figure 9 shown, according to some example embodiments, method 600 may include one or more of operations 801, 802, 901, and 902. As referenced Figure 8 above, method operation 801 may be performed as part of method operation 630 (e.g., a precursor task, subtask, or part), where image analysis module 520 performs an image evaluation of the image. In method operation 801, image analysis module 520 extracts one or more visual features from the image received from the seller.

[0116] As referenced Figure 8 above, method operation 802 may be performed after method operation 801. In method operation 802, image analysis module 520 identifies the display type for displaying the item within the image. Identifying the display type for displaying the item within the image may be based on one or more visual features extracted from the image received from the seller. In some example embodiments, method operations 801 and 802 are performed as part of image analysis 420 as referenced Figure 4 above.

[0117] Method operation 901 may be performed after method operation 802. In method operation 901, image analysis module 520 determines an image score value for the received image. The image score value for the image may be determined based on the display type for displaying the item within the image.

[0118] Method operation 902 may be performed after method operation 901. In method operation 902, the image analysis module 520 determines the likelihood that a buyer viewing the image engages in an interest display action (e.g., clicks) related to the image. Determining the likelihood that a buyer viewing the image engages in an interest display action related to the image may be based on image analysis (or the result of image analysis, e.g., the image score value of the image) and user behavior data (e.g., the result of user behavior analysis).

[0119] As Figure 10 shown, the network node 600 may include method operation 1001. Method operation 1001 may be performed as part of method operation 630 (e.g., a precursor task, a subtask, or a part), where the image analysis module 520 performs an image evaluation of the image. In method operation 1001, the image analysis module 520 determines that the image is superior to different images. Determining that the image is superior to different images may be based on a comparison between the image (e.g., a first image received from a first seller) and different images (e.g., a second image received from the first seller or a second image received from a different seller).

[0120] In certain example embodiments, to compare the image with different images, the image analysis module 520 uses one or more attribute comparison rules to perform a comparison of one or more image attribute-value pairs of the image and one or more corresponding image attribute-value pairs of the different images. For example, the image analysis module 520 may identify the corresponding values of the "clarity" attribute of the first image and the second image. Based on applying an attribute comparison rule that specifies a higher grading for the attribute value corresponding to the "clarity" attribute, the image analysis module 520 may determine which of the first image and the second image has an extremely high grading value for the "clarity" attribute. Thus, the image analysis module 520 may identify the image with the higher grading value for the "clarity" attribute as the better image.

[0121] According to different examples, the image analysis module 520 can identify corresponding values of the "display type" attribute of the first image and the second image. Based on applying an attribute comparison rule that prescribes that the attribute values corresponding to the "display type" attribute are graded higher, the image analysis module 520 can determine which of the first image and the second image has an extremely high graded value for the "display type" attribute. An attribute comparison rule that prescribes that the attribute values corresponding to the "display type" attribute (e.g., "human", "mannequin", or "flat display") are graded higher (e.g., "human" is graded higher than "mannequin", or "mannequin" is graded higher than "flat display") can be generated during the analysis of user behavior data. Therefore, the image analysis module 520 can identify an image having a higher graded value for the "display type" attribute as a better image.

[0122] As Figure 11 shown, according to some example embodiments, method 600 may include one or more of operations 1101, 1102, and 1103. Method operation 1101 may be performed as part of method operation 1001 (e.g., a precursor task, a subtask, or a part), where the image analysis module 520 determines that the image is better than a different image. In method operation 1101, the image analysis module 520 determines that the image is better than a different image based on the image being of a human display type (e.g., using a real model to display the item shown in the image).

[0123] Method operation 1102 may be performed as part of method operation 1001 (e.g., a precursor task, a subtask, or a part), where the image analysis module 520 determines that the image is better than a different image. In method operation 1102, the image analysis module 520 determines that the image is better than a different image based on the image being of a mannequin display type (e.g., using a mannequin to display the item shown in the image).

[0124] Method operation 1103 may be performed as part of method operation 1001 (e.g., a precursor task, a subtask, or a part), where the image analysis module 520 determines that the image is better than a different image. In method operation 1103, the image analysis module 520 determines that the image is better than a different image based on the image being of a flat display type (e.g., flatly displaying the item shown in the image without using a real model or a mannequin).

[0125] As Figure 12As shown, network node 600 may include method operation 1201. Method operation 1201 may be performed as part of method operation 1001 (e.g., a precursor task, a subtask, or a part thereof), where image analysis module 520 determines that the image is superior to different images. In method operation 1201, image analysis module 520 determines that the image is superior to different images based on the image having a higher grading value than the different images. Image analysis module 520 may grade a plurality of images received from a user (e.g., a seller) according to one or more grading rules. The result of the grading may identify a specific order of the images.

[0126] In some example embodiments, grading may be performed on the image and the different images within the category to which the image and the different images belong based on the value of the corresponding display type as described above. For example, if both the image and the different images use the human display type to display the items shown in the image, the image and the different images may be graded within the human image category. In some example embodiments, the images within a category may be graded based on the image score value of the image within the specific category. Figure 4 In some example embodiments, the image and the different images may be globally graded based on the image request score value (e.g., graded regardless of the category to which the image belongs). In some example embodiments, image analysis module 520 determines the image request score value of the image using a specific formula based on the classification of the image.

[0127] For example, the user preferences for each PMF are P, M, and F (corresponding to the values of 0.37, 0.31, 0.31 in Table 4 above). The trust score value is C. If the image is classified as P, the formula is:

[0128] P x C + M x (1 - C) / 2 + F x (1 - C) / 2.

[0129] Similarly, if the image is classified as M, the formula is:

[0130] P x (1 - C) / 2 + M x C + F x (1 - C) / 2.

[0131] Similarly, if the image is classified as F, the formula is:

[0132] P x (1 - C) / 2 + M x (1 - C) / 2 + F x C.

[0133]

[0134] ​For example, based on the analysis of user behavior data, the reference score values for images of different display types can be as follows: P = 0.37, M = 0.31, and F = 0.31. If a given image A is classified as a P type with a trust score value of 0.7, then the image request score value of image A is equal to:

[0135] P x C + M x (1 - C) / 2 + F x (1 - C) / 2 = 0.37 x 0.7 + 0.31 x (1 - 0.7) / 2 + 0.31 x (1 - 0.7) / 2.

[0136] The image request score value of image A can be further combined with the low-level quality score value of image A to calculate the final score value. In some example embodiments, the final score value is used to determine the global grading of multiple images.

[0137] In some examples, the combination of the image request score value and the low-level quality score value can include multiplying the image request score value of image A and the low-level quality score value of image A to calculate the final score value of image A. In some examples, the combination of the image request score value of image A and the low-level quality score value of image A can include assigning specific weights to the image request score value of image A and the low-level quality score value of image A to produce a weighted image request score value of image A and a weighted low-level quality score value of image A. The combination further includes adding the weighted image request score value of image A and the weighted low-level quality score value of image A to calculate the final score value of image A. In some example embodiments, specific weights can be selected during the analysis of user behavior data. Some or all of the score values attributed to an image (e.g., trust score value, image score value, or final score value) can be stored in one or more records of a database (e.g., database 126).

[0138] As Figure 13 shown, according to some example embodiments, method 600 can include one or more of operations 1301, 1302, and 1303. Method operation 1301 can be performed as part of method operation 1001 (e.g., a predecessor task, a subtask, or a part), where the image analysis module 520 determines that the image is superior to different images. In method operation 1301, the image analysis module 520 determines that the image is superior to different images based on the image having a higher grading value than the different images, where the image and the different images are of human display types.

[0139] Method operation 1302 can be performed as part of method operation 1001 (e.g., a predecessor task, a subtask, or a part thereof), where the image analysis module 520 determines that the image is superior to different images. In method operation 1302, the image analysis module 520 determines that the image is superior to different images based on the image having a higher grading value than the different images, where the image and the different images are human model display types.

[0140] Method operation 1303 can be performed as part of method operation 1001 (e.g., a predecessor task, a subtask, or a part thereof), where the image analysis module 520 determines that the image is superior to different images. In method operation 1303, the image analysis module 520 determines that the image is superior to different images based on the image having a higher grading value than the different images, where the image and the different images are flat display types.

[0141] As Figure 14 shown, according to some example embodiments, method 600 can include one or more of operations 1401, 1402, and 1403. Method operation 1401 can be performed as part of method operation 640 (e.g., a predecessor task, a subtask, or a part thereof), where the output module 540 generates an output referring to the image. In method operation 1401, the output module 540 generates feedback referring to the image. The feedback can include an image evaluation result of the image (e.g., evaluating the image to determine the likelihood that a user viewing the image engages in an interest display action related to the image or the item), an explanation of the image evaluation result, a report of the image evaluation, a comparison of multiple images received from a seller, a comparison of an image submitted by a seller and images submitted by other sellers (e.g., based on the image score value or image grading value of the corresponding images), examples of good images and bad images, or a suitable combination thereof.

[0142] Method operation 1402 can be performed as part of method operation 640 (e.g., a predecessor task, a subtask, or a part thereof), where the output module 540 generates an output referring to the image. In method operation 1402, the output module 540 generates a suggestion referring to the image. The suggestion can include a suggestion for an improved image of the item. The improved image of the item can increase the likelihood of obtaining a desired result from a user (e.g., a potential buyer), such as the likelihood of engaging in an interest display activity related to the image or the item shown in the image (e.g., clicking on the image or purchasing the item). In some examples, the suggestion can include a description of changes to be made to multiple features (e.g., image attributes) of the image received from the seller.

[0143] The suggestions can include the image evaluation result of the image (e.g., the image is evaluated to determine the likelihood that a user who sees the image engages in an interest display action related to the image or the item), the interpretation of the image evaluation result, the comparison of multiple images received from the seller, the report of the image evaluation, the comparison of the image submitted by the seller and the images submitted by other sellers (e.g., based on the image score value or image grading value of the corresponding image), the suggestion of selecting a more effective display type to display the item in the image, the suggestion of modifying the value of one or more other image attributes (e.g., better lighting, white background, professional photography, fewer items shown in the image or better image composition), a set of guidelines to assist the seller in how to improve the image showing the item (e.g., cost-benefit analysis of different image improvement options), examples of good and bad images, or a suitable combination thereof.

[0144] Method operation 1403 can be performed as part of method operation 640 (e.g., a precursor task, a subtask, or a part), where output module 540 generates an output referring to the image. In method operation 1403, display module 540 generates a set of guidelines that assist the seller in selecting an image that may result in a desired response from the buyer. In some example embodiments, the group of intelligence can describe how to generate or select high-quality images that facilitate increasing the sales volume of the item described in the image. The guidelines can be provided to (e.g., displayed to) the seller of the item, such as on an e-commerce website where the seller can list or sell their items. In some examples, the guidelines can be provided to the seller before the seller transmits (e.g., uploads) the image to the e-commerce website. In certain examples, the guidelines can be provided to the seller after the seller transmits (e.g., uploads) the image to the e-commerce website. The set of guidelines can be customized for a specific seller based on the image evaluation result received from the specific seller. For example, when image analysis module 520 completes the image evaluation of the image received from the seller and determines that the image may require improvement, output module 540 generates a customized set of guidelines that assist the seller in selecting an image that may result in a desired response from the buyer. Communication module 550 can display the set of guidelines to the seller via the seller's device.

[0145] In certain example embodiments, output module 540 may determine what type of output to generate based on the image score value of an image received from a seller. For example, image analysis module 520 may determine that the image depicts a clothing item for display in a P type based on the visual features of the extracted image. Based on the image displayed in the P type, image analysis module 520 may assign a relatively high image score value to the image (compared to other images in M type or F type). Output module 540 may determine that the output referring to the image may include feedback referring to the image (e.g., feedback on how the image compares to other images submitted by other sellers (e.g., actively)), but may not include suggestions for improving the image based on an image that already has a relatively high image score value.

[0146] According to various example embodiments, one or more of the methods described herein may facilitate the evaluation of images depicting items during online sales. Additionally, one or more of the methods described herein may facilitate providing improved images depicting items for online sales. Thus, one or more of the methods described herein may facilitate improving the sales of items depicted in the images.

[0147] When considering these effects in general, one or more of the methods described herein may eliminate the need for certain workloads or resources that would otherwise be involved in evaluating images of items for online sales. By one or more of the methods described herein, the effort exerted by providers of such images (e.g., sellers) in evaluating these images may be reduced. The computing resources used by one or more machines, databases, or devices (e.g., in network environment 300) may be similarly reduced. Examples of such computing resources include processor cycles, network traffic, memory usage, data storage capacity, power consumption, and cooling capacity.

[0148] Example mobile device

[0149] Figure 15is a block diagram showing a mobile device 1500 according to an example embodiment. The mobile device 1500 may include a processor 1502. The processor 1502 may be any of a variety of different types of commercially available processors 1502 suitable for the mobile device 1500 (e.g., an XScale architecture microprocessor, a microprocessor without interlocked pipeline stages (MIPS) architecture processor, or another type of processor 1502). A memory 1504 (e.g., random access memory (RAM), flash memory, or other type of memory) can generally be accessed by the processor 1502. The memory 1504 may be adapted to store an operating system (OS) 1506 and application programs 1508, such as a mobile location-enabled application that can provide LBS to a user. The processor 1502 may be directly or via a suitable intermediate hardware connection connected to a display 1510 and connected to one or more input / output (I / O) devices 1512, such as a keyboard, a touchpad sensor, a microphone, etc. Similarly, in some embodiments, the processor 1502 may be connected to a transceiver 1514 that interacts with an antenna 1516. Depending on the nature of the mobile device 1500, the transceiver 1514 may be configured to transmit and receive cellular network signals, wireless data signals, or other types of signals via the antenna 1516. Additionally, in some configurations, a GPS receiver 1518 may also utilize the antenna 1516 to receive GPS signals.

[0150] Modules, Components, and Logic

[0151] Certain embodiments are described herein as including logic or a plurality of components, modules, or mechanisms. A module may constitute a software module (e.g., (1) code implemented on a non-transitory machine-readable medium, or (2) code implemented in a transmitted signal) or a hardware-implemented module. A hardware-implemented module is a tangible unit capable of performing certain operations and can be configured or arranged in a particular physical manner. In an example embodiment, one or more computer systems (e.g., a stand-alone computer system, a client computer system, or a server computer system) or one or more processors 1502 may be configured by software (e.g., an application or an application portion) to operate as a hardware module that performs certain operations described herein.

[0152] In various embodiments, hardware-implemented modules can be implemented mechanically or electronically. For example, a hardware-implemented module can include dedicated circuitry or logic that is permanently configured to perform certain operations (e.g., a hardware-implemented module can be a dedicated processor such as a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC)). A hardware-implemented module can also include programmable logic or circuitry that is temporarily configured by software to perform certain operations (e.g., programmable logic or circuitry included in a general-purpose processor 1502 or other programmable processor 1502). It will be appreciated that the decision to implement a hardware-implemented module mechanically, in dedicated and permanently configured circuitry, or in temporarily configured circuitry (e.g., configured by software) can be driven by cost and time considerations.

[0153] Accordingly, the phrase "hardware-implemented module" should be understood to include a tangible entity that is physically constructed, permanently configured (e.g., hardwired) or temporarily or transiently configured (e.g., programmed) to operate in a particular manner to perform the particular operations described herein. Considering embodiments in which a hardware-implemented module is temporarily configured (e.g., programmed), each hardware-implemented module need not be configured or instantiated at any given time. For example, in the case where a hardware-implemented module includes a general-purpose processor 1502 configured by software, the general-purpose processor 1502 can be configured to be respective different hardware-implemented modules at different times. Thus software can configure the processor 1502 to, for example, constitute a certain hardware-implemented module at one moment and a different hardware-implemented module at a different moment.

[0154] Hardware-implemented modules can provide information to, and receive information from, other hardware-implemented modules. Accordingly, the described hardware-implemented modules can be regarded as being communicatively coupled. In cases where multiple such hardware-implemented modules are present simultaneously, communication can be achieved via signal transmission (e.g., on appropriate circuits and buses connecting the hardware-implemented modules). In embodiments in which multiple hardware-implemented modules are configured or instantiated at different times, such communication between the hardware-implemented modules can be achieved, for example, via the storage and retrieval of information in a memory structure accessible to the multiple hardware-implemented modules. For example, one hardware-implemented module can perform an operation and store the output of that operation in a storage device communicatively coupled to the hardware-implemented module. Then another hardware-implemented module can access the storage device at a later time to retrieve and process the stored output. Hardware-implemented modules can also initiate communication with input or output devices and can perform operations on resources (e.g., collections of information).

[0155] The various operations of the example methods described herein may be performed, at least in part, by one or more processors 1502 that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors 1502 may constitute processor-implemented modules that operate to perform one or more operations or functions. In some example embodiments, a "module" as used herein includes processor-implemented modules.

[0156] Similarly, the methods described herein may be at least partially implemented by processors. For example, at least some operations of the method may be performed by one or more processors 1502 or processor-implemented modules. The execution of certain operations may be distributed among one or more processors 1502 or processor-implemented modules and not just reside on a single machine but be arranged among multiple machines. In some example embodiments, one or more processors 1502 or processor-implemented modules may be located at a single location (e.g., in a home environment, an office environment, or a server farm), while in other embodiments, one or more processors 1502 or processor-implemented modules may be distributed among multiple locations.

[0157] One or more processors 1502 may also operate to support the execution of relevant operations in a "cloud computing environment" or as the execution of relevant operations of "software as a service" (SaaS). For example, at least some operations may be performed by a group of computers (e.g., machines including processors) that are accessible via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., application programming interfaces (APIs)).

[0158] Electronic devices and systems

[0159] Example embodiments may be implemented in digital electronic circuitry or in computer hardware, firmware, software, or combinations thereof. Example embodiments may be implemented using a computer program product, such as a computer program tangibly embodied in an information carrier, the information carrier being, for example, a machine-readable medium executable by a data processing apparatus or a machine-readable medium for controlling the operation of a data processing apparatus, the data processing apparatus being, for example, a programmable processor 1502, a computer, or multiple computers.

[0160] A computer program may be written in any form of programming language, including compiled or interpreted languages, and may be deployed in any form, including as a stand-alone program or as modules, subroutines, or other units suitable for use in a computing environment. A computer program may be configured to be executed on one computer or on multiple computers located at one location or on multiple computers distributed among multiple locations and interconnected by a communication network.

[0161] In an example embodiment, operations may be performed by one or more programmable processors 1502 executing a computer program to perform functions by operating on input data and generating output. Method operations may also be performed by dedicated logic circuitry (e.g., a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC)), and the apparatus of the example embodiment may be implemented as dedicated logic circuitry.

[0162] A computing system may include a client and a server. The client and the server are generally remote from each other and typically interact via a communication network. The relationship of the client and the server is created by computer programs running on the respective computers and having a client-server relationship with each other. In embodiments using a programmable computing system, it will be appreciated that both the hardware architecture and the software architecture need to be considered. Specifically, it will be appreciated that implementing a particular function in permanently configured hardware (e.g., an ASIC), in temporarily configured hardware (e.g., a combination of software and a programmable processor 1502), or in a combination of permanently configured and temporarily configured hardware may be a design choice. The following are the hardware architectures (e.g., machines) and software architectures that may be deployed in various example embodiments.

[0163] Example Machine Architectures and Machine-Readable Media

[0164] Figure 16 shows components of a machine 1600 capable of reading instructions 1624 from a machine-readable medium 1622 (e.g., a non-transitory machine-readable mechanism, a machine-readable storage medium, a computer-readable storage medium, or any suitable combination thereof) and performing any one or more of the methods discussed herein, in whole or in part. Specifically, Figure 16 shows an example form of a machine 1600 in the form of a computer system, in which instructions 1624 (e.g., software, a program, an application, an applet, an app, or other executable code) for causing the machine 1600 to perform any one or more of the methods discussed herein may be executed, in whole or in part.

[0165] In an alternative embodiment, machine 1600 operates as a stand-alone device or can be connected to (e.g., networked directly to) other machines. In a networked deployment, machine 1600 can operate in the capacity of a server machine or a client machine in a server-client network environment, or as a peer machine in a distributed (e.g., peer-to-peer) network environment. Machine 1600 can be a server computer, a client computer, a personal computer (PC), a tablet computer, a laptop computer, a netbook, a cellular phone, a smartphone, a set-top box (STB), a personal digital assistant (PDA), a web device, a network router, a network switch, a bridge, or any machine capable of executing instructions 1624 sequentially or otherwise, where the instructions 2124 specify actions to be taken by that machine. Further, although only a single machine is shown, the term "machine" shall also be taken to include any collection of machines that individually or jointly execute instructions 1624 to perform any one or more of the methods discussed herein.

[0166] Machine 1600 includes a processor 1602 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a radio frequency integrated circuit (RFIC), or any suitable combination thereof), a main memory 1604, and a static memory 1606 that are configured to communicate with each other via a bus 1608. Processor 1602 can include microcircuits that can be temporarily or permanently configured by some or all of instructions 1624 such that processor 1602 can be configured to perform any one or more of the one or more methods described herein. For example, a collection of one or more microcircuits of processor 1602 can be configured to perform one or more modules (e.g., software modules) described herein.

[0167] Machine 1600 may also include a graphical display 1610 (e.g., a plasma display panel (PDP), a light emitting diode (LED) display, a liquid crystal display (LCD), a projector, a cathode ray tube (CRT), or any other display capable of displaying graphics or video). Machine 1600 may also include an alphanumeric input device 1612 (e.g., a keyboard or keypad), a cursor control device 1614 (e.g., a mouse, a touchpad, a trackball, a joystick, a motion sensor, an eye tracking device, or other positioning instrument), a storage unit 1616, an audio generation device 1618 (e.g., a sound card, an amplifier, a speaker, a headphone jack, or any suitable combination thereof), and a network interface device 1620.

[0168] The storage unit 1616 includes a machine-readable medium 1622 (such as a tangible and non-transitory machine-readable storage medium) in which instructions 1624 are stored, and the instructions 2124 implement any one or more of the methods or functions described herein. The instructions 1624 may also reside, fully or at least partially, within the main memory 1604, within the processor 1602 (such as within the processor's cache), or within both, before or during execution by the machine 1600. Accordingly, the main memory 1604 and the processor 1602 can be regarded as machine-readable media (such as tangible and non-transitory machine-readable media). The instructions 1624 can be sent or received via the network 1626 through the network interface device 1620. For example, the network interface device 1620 can transmit the instructions 1624 using any one or more transport protocols (such as the Hypertext Transfer Protocol (HTTP)).

[0169] In some example embodiments, the machine 1600 can be a portable computing device (such as a smart phone or a tablet computer), and has one or more additional input components 1630 (such as sensors or meters). Examples of such input components 1630 include image input components (such as one or more cameras), audio input components (such as microphones), orientation input components (such as compasses), position input components (such as Global Positioning System (GPS) receivers), orientation components (such as one or more accelerometers), altitude detection components (such as altimeters), and gas detection components (such as gas sensors). The inputs obtained by any one or more of these input components are accessible and available for use by any of the modules described herein.

[0170] As used herein, the term "memory" refers to a machine-readable medium capable of storing data either temporarily or permanently, and may be regarded as including, but not limited to, random access memory (RAM), read-only memory (ROM), buffer memory, flash memory, and cache memory. Although the machine-readable medium 1622 is shown as a single medium in the exemplary embodiment, the term "machine-readable medium" should be regarded as including a single medium or multiple media (e.g., a centralized or distributed database, or associated cache and servers) capable of storing instructions. The term "machine-readable medium" should also be regarded as including any combination of a medium or multiple media capable of storing the instructions 1624 for execution by the machine 1600 such that when the instructions 1624 are executed by one or more processors (e.g., processor 1602) of the machine 1600, the machine 1600 performs, in whole or in part, any one or more of the methods described herein. Thus, "machine-readable medium" refers to a single storage device or apparatus, as well as a cloud-based storage system or a storage network including multiple storage devices or apparatuses. Accordingly, the term "machine-readable mechanism" should be regarded as including one or more tangible (e.g., non-transitory) data repositories of solid-state memory, optical media, magnetic media, or any suitable combination thereof. A machine-readable medium may also include transitory media, such as a signal or carrier medium, such as an electromagnetic signal, an electrical signal, an optical signal, or an acoustic signal carrying the machine-readable instructions.

[0171] In this specification, plural instances may implement components, operations, or structures described as a single instance. While various operations of one or more methods are illustrated and described as separate operations, one or more of the various operations may be performed concurrently, and the operations need not be performed in the order shown. Structures and functions that are shown as separate components in exemplary configurations may be implemented as a combined structure or component. Similarly, structures and functions that are shown as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements are within the scope of the subject matter here.

[0172] Certain embodiments are described herein as including logic or multiple components, modules, or mechanisms. A module may include a software module (e.g., code stored or implemented on a machine-readable medium or a transmission medium), a hardware module, or any suitable combination thereof. A "hardware module" is a tangible (e.g., non-transitory) unit capable of performing certain operations and may be physically configured or arranged in a certain manner. In various exemplary embodiments, one or more computer systems (e.g., a stand-alone computer system, a client computer system, or a server computer system) or one or more hardware modules of a computer system (e.g., a processor or a group of processors) may be configured by an element (e.g., an application or a portion of an application) as a hardware module to perform specific operations described herein.

[0173] In some embodiments, the hardware module can be implemented mechanically, electronically, or any suitable combination thereof. For example, the hardware module can include dedicated circuits or logic that are permanently configured to perform specific operations. For example, the hardware module can be a dedicated processor, such as a field programmable gate array (FPGA) or an ASIC. The hardware module can also include programmable logic or circuits that are temporarily configured by software to perform specific operations. For example, the hardware module can include software contained in a general-purpose processor or other programmable processor. It should be understood that the decision to implement mechanically, with dedicated and permanently configured circuits, or with circuits configured sub-optimally (e.g., configured by software) can be for cost and time considerations.

[0174] Accordingly, the phrase "hardware module" should be understood to cover a tangible entity, and such that the tangible entity is physically constructed, permanently configured (e.g., hardwired) or temporarily configured (e.g., programmed) to operate in a particular manner or to perform the specific operations described herein. As used herein, "hardware-implemented module" refers to a hardware module. Considering embodiments of a hardware module that are temporarily configured (e.g., programmed), it is not necessary to configure or instantiate each of the hardware modules at any given moment. For example, if a hardware module includes a general-purpose processor that is configured by software to be a dedicated processor, the general-purpose processor can be configured at different times to be respectively different dedicated processors (e.g., including different hardware modules). Thus, software (e.g., a software module) can configure one or more processors, for example, to constitute a particular hardware module at one moment and a different hardware module at another moment.

[0175] A hardware module can provide information to and receive information from other hardware modules. Accordingly, the described hardware modules can be considered to be communicatively coupled. If multiple hardware modules are present simultaneously, communication can be achieved through signal transmission between two or more of the hardware modules (e.g., via appropriate circuits and buses). In embodiments where multiple hardware modules are configured or instantiated at different times, communication between such hardware modules can be achieved, for example, by storing and retrieving information in a memory structure accessible to the multiple hardware modules. For example, one hardware module can perform an operation and store the output of the operation in a storage device to which it is communicatively coupled. Another hardware module can then later access the memory device to retrieve and process the stored output. A hardware module can also initiate communication with an input or output device and be capable of operating on resources (e.g., a collection of information).

[0176] The various operations of the example methods described herein can be performed, at least in part, by one or more processors temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors constitute processor-implemented modules that operate to perform one or more of the operations or functions described herein. As used herein, a "processor-implemented module" refers to a hardware module implemented using one or more processors.

[0177] Similarly, the methods described herein can be at least in part processor-implemented, where the processor is an example of hardware. For example, at least some of the operations of the method can be performed by one or more processors or processor-implemented modules. As used herein, a "processor-implemented module" refers to a hardware module where the hardware includes a processor. Additionally, one or more processors can also operate to perform the relevant operations in a "cloud computing" environment or as "software as a service" (SaaS). For example, at least some of the operations can be performed by a group of computers (as an example of machines including processors), and these operations can be accessed via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., application programming interfaces (APIs)).

[0178] Performing a particular operation can be distributed among one or more processors that are not only located on a single machine but also deployed across multiple machines. In some example embodiments, one or more processors or processor-implemented modules can be located in a single geographical location (e.g., in a home environment, an office environment, or a server farm). In other example embodiments, one or more processors or processor-implemented modules can be distributed across multiple geographical locations.

[0179] Some portions of the subject matter discussed herein can be presented in an algorithmic or symbolic representation of operations on data stored as bits or binary digital signals within a machine memory (e.g., a computer memory). Such algorithmic or symbolic representations are examples of techniques used by those of ordinary skill in the data processing arts to convey the substance of their work to others skilled in the art. As used herein, an "algorithm" is a self-consistent sequence of operations or similar processing that results in a desired outcome. In this context, algorithms and operations involve physical manipulations of physical quantities. Usually, but not necessarily, such quantities can take the form of electrical, magnetic, or optical signals capable of being stored, accessed, transmitted, combined, compared, or otherwise manipulated by a machine. Sometimes, mainly for reasons of common usage, it is convenient to use words such as "data," "content," "bits," "values," "elements," "symbols," "characters," "items," "numbers," "digits," etc. to refer to such signals. However, these words are merely convenient labels and should be associated with the appropriate physical quantities.

[0180] Unless otherwise specifically stated, discussions herein using terms such as "processing", "computing", "operating", "determining", "rendering", "displaying", etc. may refer to actions or processes of a machine (such as a computer) that manipulates or transforms data represented as physical (such as electronic, magnetic, or optical) quantities within one or more memories (such as volatile memory, non-volatile memory, or any combination thereof), registers, or other machine components that receive, store, transmit, or display information. Additionally, unless otherwise specifically stated, as is common in patent literature, the term "a" or "an" herein is used to include one or more than one instance. Finally, as used herein, unless otherwise specified, the conjunction "or" refers to a non-exclusive "or".

[0181] Industrial Applicability

[0182] The present invention disclosed herein has broad industrial applicability, such as image analysis, data processing, and human-machine interaction.

Claims

1. A system for image evaluation, comprising: a memory storing a database including one or more analysis results of user behavior data related to a plurality of test images; one or more hardware processors; a receiver module implemented by the one or more processors and configured to: access one or more analysis results of user behavior data; and receive an image of an item from a user device; an image analysis module implemented by the one or more processors and configured to perform an evaluation of the received image based on the one or more analysis results of the user behavior data, performing the evaluation including: calculating a score value of the received image based on the one or more analysis results of the user behavior data and values of one or more image attributes of the received image, and determining a likelihood that a user will engage in a required user behavior related to the received image based on the score value of the received image; and an output module implemented by the one or more processors and configured to: generate an output for the user device based on the evaluation of the received image, the output referring to the received image and indicating the likelihood that a user will engage in a required user behavior related to the received image.

2. The system according to claim 1, wherein the score value for the received image is a first score value, and wherein the image analysis module is further configured to: compare the received image with another image of another similar item based on the first score value and a second score value corresponding to the other image, and generate a ranking of the received image related to the other image of the other similar item.

3. The system according to claim 2, wherein the output for the user device further indicates the ranking of the received image relative to the other images.

4. The system according to claim 2, wherein the image analysis module is further configured to classify the received image into an image category based on values of one or more image attributes of the received image, the image category including the other image, and wherein the received image and the other image are ranked within the image category.

5. The system according to claim 2, wherein the image analysis module is further configured to: access values of one or more image attributes of the other image; and calculate a second score value corresponding to the other image based on the values of the one or more image attributes of the other image.

6. The system according to claim 1, further comprising: a communication module configured to transmit a communication message to the user device, the communication message including an output generated with reference to the received image.

7. The system according to claim 1, wherein the image analysis module is further configured to: extract one or more visual features from the received image, and identify values of one or more attributes of the received image based on an analysis of the one or more visual features.

8. The system according to claim 1, wherein one or more attributes of the received image include a display type of an item displayed within the received image, and wherein calculating the score value of the received image is based on the value of the display type.

9. The system according to claim 1, further comprising a behavior analysis module configured to: Access the user behavior data; Analyze the user behavior data; Generate one or more analysis results of the user behavior data; and Store one or more analysis results of the user behavior data in a database.

10. A computer-implemented method for image evaluation, comprising: Accessing one or more analysis results of user behavior data related to a plurality of test images; Receiving an image of an item from a user device; Performing an evaluation of the received image based on one or more analysis results of the user behavior data, wherein performing the evaluation includes: Calculating a score value of the received image based on one or more analysis results of the user behavior data and values of one or more image attributes of the received image, and Determining a likelihood that a user will engage in a required user behavior related to the received image based on the score value of the received image; And Generating an output for the user device based on the evaluation of the received image, the output referring to the received image and indicating a likelihood that a user will engage in a required user behavior related to the received image.

11. The computer-implemented method according to claim 10, further comprising: Classifying the received image into an image category based on values of one or more image attributes of the received image, the image category including other images.

12. The computer-implemented method according to claim 11, wherein the score value of the received image is further based on a trust score value of the received image, the trust score value measuring a confidence level of classifying the received image into the category to which the received image belongs.

13. The computer-implemented method according to claim 11, wherein the received image and other images are classified in the category based on recognition values of display type attributes of the received image and other images.

14. The computer-implemented method according to claim 13, wherein the score value of the received image is further based on a combination of a trust score value of the received image and a low-level quality score value of the received image, the trust score value measuring classifying the received image into a category corresponding to a value of the display type attribute of the received image, the low-level quality score value measuring the quality of the received image based on another value of another image attribute of the received image.

15. The computer-implemented method according to claim 10, wherein performing the evaluation of the received image includes determining that the received image is superior to different images.

16. The computer-implemented method according to claim 15, wherein determining that the received image is superior to different images is based on the received image being of a human display type.

17. The computer-implemented method according to claim 15, wherein determining that the received image is superior to different images is based on the received image being of a mannequin display type.

18. The computer-implemented method according to claim 15, wherein determining that the received image is superior to different images is based on the received image being of a flat display type.

19. The computer-implemented method according to claim 15, wherein determining that the received image is superior to a different image is based on the received image having a higher rank value than the different image.

20. A machine-readable medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations including: accessing one or more analysis results of user behavior data associated with a plurality of test images; receiving an image of an item from a user device; performing an evaluation of the received image based on one or more analysis results of the user behavior data, the performing the evaluation including: calculating a score value of the received image based on one or more analysis results of the user behavior data and values of one or more image attributes of the received image, and determining a likelihood that a user will engage in a required user behavior associated with the received image based on the score value of the received image; and generating an output for the user device based on the evaluation of the received image, the output referring to the received image and indicating the likelihood that a user will engage in a required user behavior associated with the received image.

21. A machine-readable medium carrying instructions that, when executed by one or more processors of a machine, cause the machine to perform the method according to any one of claims 10 to 19.

Images