Artificial intelligence-based blind date matching system based on face image

Abstract

The present invention provides an artificial intelligence-based blind date matching system based on a face image, the system comprising a matching server for generating matching information by storing and analyzing a face image transmitted from a user terminal, and providing the generated matching information to the user terminal. The matching server includes: an information collection unit for receiving and collecting user information and the face image from the user terminal; a feature extraction unit for generating a face template by extracting face feature information through an artificial intelligence algorithm on the basis of the face image collected by the information collection unit; a storage unit for databasing and storing the face image and face template information together with the user information; a matching information generation unit for generating matching information by comparing the face template information about a user of the user terminal, which has transmitted a matching request signal, with face template information about another user stored in the storage unit when the matching request signal is transmitted from the user terminal; and an information providing unit for transmitting the matching information generated by the matching information generation unit to the user terminal that has transmitted the matching request signal.

Description

AI-based blind date matching system based on facial images

[0001] The present invention relates to an AI-based blind date matching system based on facial images. More specifically, it relates to an AI-based blind date matching system that compares and analyzes the faces of multiple users stored on a server using AI deep learning to find and introduce similar partners.

[0002] Examples of arranged encounters for dating include blind dates, meetups, and even matchmaking for marriage. Arranged encounters refer to encounters between men and women seeking a romantic relationship, facilitated by a matchmaker. While these encounters were typically conducted through acquaintances, relatives, or matchmakers, they are now increasingly being commercialized through professional matchmaking agencies.

[0003] Matchmaking companies store specific information about each registered male and female user in a computerized system, match members with suitable conditions, and generate additional revenue when they marry. To find the right partner through a vast database of member information, technology has been developed to search for members based on various criteria, such as location, age, body type, and occupation. Matching algorithms have been improved to recommend carefully selected members.

[0004] However, these matching algorithms are generally used in a way that users input desired conditions and find and match the counterparty that best matches the input conditions.

[0005] Since matching information is generated mainly based on conditions through these matching algorithms, it is very difficult to develop mutual feelings for each other, and it is very rare for it to lead to actual dating or marriage.

[0006] It is known that both men and women are instinctively attracted to people who resemble each other in relationships between the sexes. Therefore, there is a need for an algorithm that matches people by finding similar people rather than a matching algorithm based on conditions. However, such a matching system is currently underdeveloped.

[0007] The present invention was invented to solve the problems of the prior art, and the purpose of the present invention is to provide an artificial intelligence-based blind date matching system that compares and analyzes the faces of multiple users stored in a server through AI deep learning to find and introduce the most similar person to them.

[0008] Another object of the present invention is to provide an artificial intelligence-based blind date matching system that can provide matching information more quickly and accurately by first selecting comparison data through separate classification criteria or restriction conditions in the process of comparing and analyzing faces of multiple users, thereby shortening the time required to generate matching information.

[0009] Another object of the present invention is to provide an artificial intelligence-based blind date matching system that can provide more accurate matching information by providing users to be matched in order of ranking, enabling the user to make a selection, and analyzing the results of the user's selection to reflect in selecting the users to be matched.

[0010] The present invention provides an artificial intelligence-based blind date matching system based on a facial image, comprising: a matching server that stores and analyzes a facial image transmitted from a user terminal to generate matching information and provides the generated matching information to the user terminal; the matching server comprises: an information collection unit that receives and collects user information and a facial image transmitted from the user terminal; a feature extraction unit that extracts facial feature information through an artificial intelligence algorithm based on the facial image collected by the information collection unit to generate a facial template; a storage unit that stores the facial image and facial template information together with user information in a database; a matching information generation unit that, when a matching request signal is transmitted from the user terminal, compares the facial template information of the user of the user terminal that transmitted the matching request signal with the facial template information of another user stored in the storage unit to generate matching information; and an information provision unit that transmits the matching information generated by the matching information generation unit to the user terminal that transmitted the matching request signal.

[0011] At this time, the matching information generation unit compares the face template information of the matching request user, who is the user of the user terminal that transmitted the matching request signal, with the face template information of another user, and extracts the user whose face template information has the highest similarity with the matching request user based on the comparison result, and matches the extracted user with the matching request user to generate the matching information.

[0012] In addition, the matching information generation unit includes a similarity calculation unit that compares the face template information of the matching request user stored in the storage unit with the face template information of another user to calculate a similarity; and a matching target selection unit that selects a user with the highest similarity based on the calculation result of the similarity calculation unit as a matching target user and generates the matching information, and the matching information may include a face image and user information of the matching target user and a similarity value for the matching target user.

[0013] In addition, the matching information generation unit further includes a primary selection unit that selects users according to the restriction conditions from among the user information stored in the storage unit, taking into account the restriction conditions transmitted from the user terminal, and the similarity calculation unit can calculate the similarity for the users selected by the primary selection unit.

[0014] Additionally, the above restrictions may include gender conditions, age conditions, and regional conditions.

[0015] In addition, the matching information generation unit further includes a classification unit that classifies a plurality of face template information stored in the storage unit into a plurality of groups according to facial features, and the similarity calculation unit can calculate a similarity by comparing the face template information of the matching request user with the face template information of another user within a group to which the face template information of the matching request user belongs among the plurality of groups classified by the classification unit.

[0016] In addition, the classification unit classifies into multiple groups for each classification criterion based on multiple classification criteria according to facial features, and when the classification criterion is transmitted from the user terminal by input of the matching request user, the similarity calculation unit can calculate the similarity within the group to which the face template information of the matching request user belongs among the multiple groups classified according to the transmitted classification criterion.

[0017] In addition, the matching target selection unit may select a plurality of matching target users in order of high similarity up to a preset standard ranking so that the matching request user can select them, and select the user selected by the matching request user as the final matching target user to generate the matching information.

[0018] In addition, the matching information generation unit further includes a selection analysis unit that analyzes the result of the matching request user selecting the final matching target user and extracts the facial feature having the highest similarity among the facial features between the matching request user and the final matching target user, and the similarity calculation unit can calculate the similarity by assigning a weight to the facial feature extracted through the selection analysis unit.

[0019] According to the present invention, the faces of multiple users stored in a server can be compared and analyzed through AI deep learning to find and introduce the most similar person to them, thereby increasing the probability of successful matching.

[0020] In addition, by first selecting comparison data through separate classification criteria or restriction conditions in the process of comparing and analyzing faces for multiple users, the time required to generate matching information can be shortened, allowing for more rapid and accurate matching information provision.

[0021] In addition, by providing users to be matched in order of ranking, enabling users to make their selections, and analyzing the results of the users' selections to reflect in the process of selecting users to be matched, it is possible to provide more accurate matching information.

[0022] FIG. 1 is a diagram conceptually illustrating the overall configuration of a blind date matching system according to one embodiment of the present invention.

[0023] FIG. 2 is a functional block diagram functionally illustrating the configuration of a matching server of a blind date matching system according to one embodiment of the present invention.

[0024] FIG. 3 is a functional block diagram functionally illustrating the configuration of a matching information generation unit of a matching server according to one embodiment of the present invention.

[0025] Figure 4 is a flowchart illustrating step-by-step the operation state of a blind date matching system according to one embodiment of the present invention.

[0026] The present invention relates to an artificial intelligence-based blind date matching system based on a facial image, and includes a matching server that stores and analyzes a facial image transmitted from a user terminal to generate matching information and provides the generated matching information to the user terminal, wherein the matching server may include an information collection unit that receives and collects user information and a facial image transmitted from the user terminal; a feature extraction unit that extracts facial feature information through an artificial intelligence algorithm based on the facial image collected by the information collection unit to generate a facial template; a storage unit that stores the facial image and facial template information together with user information in a database; a matching information generation unit that, when a matching request signal is transmitted from a user terminal, compares the facial template information of the user of the user terminal that transmitted the matching request signal with the facial template information of another user stored in the storage unit to generate matching information; and an information provision unit that transmits the matching information generated by the matching information generation unit to the user terminal that transmitted the matching request signal.

[0027] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. First, when assigning reference numerals to components in each drawing, it should be noted that, where possible, identical components are assigned the same reference numerals even if they appear in different drawings. Furthermore, when describing the present invention, detailed descriptions of known related structures or functions will be omitted if they are deemed to obscure the gist of the present invention.

[0028]

[0029] FIG. 1 is a diagram conceptually illustrating the overall configuration of a blind date matching system according to one embodiment of the present invention, FIG. 2 is a functional block diagram functionally illustrating the configuration of a matching server of a blind date matching system according to one embodiment of the present invention, FIG. 3 is a functional block diagram functionally illustrating the configuration of a matching information generation unit of a matching server according to one embodiment of the present invention, and FIG. 4 is a flowchart step-by-step illustrating the operating state of a blind date matching system according to one embodiment of the present invention.

[0030] A blind date matching system according to one embodiment of the present invention is configured to include a matching server (100) that is connected to a plurality of user terminals (10) for communication.

[0031] The matching server (100) can receive, store and analyze facial images from multiple user terminals (10), and, based on the analysis results, create matching information for the most similar user and provide the matching information to the user terminal (10).

[0032] The user terminal (10) is connected to the matching server (100) via a communication network. When a matching request signal is transmitted to the matching server (100) via the user terminal (10), the matching server (100) creates matching information for a user who most resembles the matching requesting user and provides the information to the user terminal (10).

[0033] The user terminal (10) may be, but is not limited to, a computer device, and may include a smart phone, a mobile phone, a navigation device, a laptop, a digital broadcasting terminal, a PDA (Personal Digital Assistant), a PMP (Portable Multimedia Player), a tablet PC, etc. For example, the user terminal (10) may communicate with the matching server (100) through a communication network using a wireless or wired communication method.

[0034] The communication network is not limited to a communication method, and may include, for example, communication methods utilizing a mobile communication network, wired Internet, wireless Internet, and broadcasting network, as well as short-range wireless communication between devices. For example, the communication network may include any one or more of a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), a broadband network (BBN), and the Internet. In addition, the communication network may include any one or more of a network topology including, but not limited to, a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree, or a hierarchical network.

[0035] According to one embodiment of the present invention, a user terminal (10) can input and transmit a matching request signal or output received matching information on a screen through a web-based program provided by a matching server (100). For example, the matching server (100) can variably utilize resources for a large number of user services based on MSA (Micro Service Architecture). This allows the same user environment to be provided on various platforms such as PCs, tablets, and mobile phones without a separate dedicated app. However, the present invention is not limited thereto, and the user terminal (10) can input and transmit a matching request signal or output received matching information on a screen through a dedicated app separately provided by the matching server (100). Here, the dedicated app may be provided through an app market of a known smartphone operating system, such as the Android Store or App Store, or may be provided directly by the matching server (100).

[0036] The matching server (100) analyzes a facial image using an artificial intelligence algorithm to generate matching information and transmits it to the user terminal (10). This matching server (10) has the same hardware configuration as a typical web server, and can include program modules that perform various functions by being implemented in various languages ​​such as C, C++, Java, Visual Basic, Visual C, etc. in terms of software. In addition, it can be implemented using a web server program that is provided in various ways according to an operating system such as DOS, Windows, Linux, Unix, Macintosh, Android, iOS, etc. on general server hardware.

[0037] A matching server (100) according to one embodiment of the present invention includes an information collection unit (110), a feature extraction unit (120), a storage unit (130), a matching information generation unit (140), and an information provision unit (150).

[0038] The information collection unit (110) receives and collects user information and facial images from multiple user terminals (10). The user information and facial images collected from the user terminals (10) are stored in a database together with the facial template information described below and stored in the storage unit (130).

[0039] The feature extraction unit (120) extracts facial feature information using an artificial intelligence algorithm based on the facial image collected by the information collection unit (110) to create a facial template.

[0040] A more detailed look at the process of generating a face template using an AI algorithm begins with a deep learning-based face detection algorithm that detects and crops face images from the entire image. This process detects face images using learning results from a large number of images. The face images are then aligned using landmarks within the face images. After the face image detection and alignment are completed, the process of extracting facial features is performed. The facial feature extraction process takes the preprocessed image as input and creates a face template. A face template is a numerical representation of the features extracted from the face image by a deep learning model. More specifically, a face template is a high-dimensional vector that quantifies key features extracted from the face image and represents a collection of data representing facial elements such as the eyes, nose, mouth, and facial contour. Because each person's face has unique features, each individual face template can be represented by a unique number, representing a unique biometric information about the individual. Therefore, face template information can be used for similarity calculations, such as comparing different face images.

[0041] When a matching request signal is transmitted from a user terminal (10), the matching information generation unit (140) compares the face template information of the user of the user terminal (10) that transmitted the matching request signal with the face template information of another user stored in the storage unit (130) to generate matching information.

[0042] The matching information generated by the matching information generation unit (140) is transmitted to the user terminal (10) that transmitted the matching request signal by the information provision unit (150).

[0043] At this time, the process of generating matching information through the matching information generation unit (140) compares the face template information of the matching request user with the face template information of another user, matches the user with the highest similarity with the matching request user, and generates matching information.

[0044] That is, the matching information generation unit (140) compares the face template information of the matching request user who sent the matching request signal with the face template information of another user, extracts the user whose face template information has the highest similarity with the matching request user based on the comparison result, and matches the extracted user with the matching request user to generate matching information.

[0045] This matching information generation unit (140) includes a similarity calculation unit (143) that calculates similarity by comparing the face template information of the matching request user stored in the storage unit (130) with the face template information of another user, and a matching target selection unit (144) that selects the user with the highest similarity based on the calculation result of the similarity calculation unit (143) as the matching target user and generates matching information.

[0046] At this time, the matching information may include a similarity value for the matched user along with the face image and user information of the matched user.

[0047] In this way, by calculating the user who most resembles the matching request user through an artificial intelligence algorithm through the matching information generation unit (140) and providing the user as matching information, the most similar person known to be instinctively most attractive can be matched, thereby improving the probability of successful matching.

[0048] In this way, in the process of selecting a matching target, if all of the user's facial images stored in the storage unit (130) are analyzed, it is impossible to filter out gender, age, region, etc. at all, and the matching server (100) may take a long time to calculate, so the matching information generation unit (140) may be equipped with a separate primary selection unit (142).

[0049] The primary selection unit (142) may be configured to initially select users according to the restriction conditions among the user information stored in the storage unit (130), taking into account the restriction conditions transmitted from the user terminal (10). At this time, the restriction conditions may include gender conditions, age conditions, and regional conditions. Thereafter, the similarity calculation unit (143) may calculate the similarity only for the users selected by the primary selection unit (142).

[0050] In this case, a restriction condition selection input screen that allows selection and input of restriction conditions may be displayed on the user terminal (10).

[0051] In this way, by limiting the range of users to be compared to within a certain range according to the constraints set by the user in the process of selecting a matching target, the operation time of the matching server (100) can be shortened, so that matching information can be generated more quickly, and the user can more accurately select the matching partner desired.

[0052] In addition, the matching information generation unit (140) may additionally include a classification unit (141) to shorten the calculation process of the artificial intelligence algorithm in the process of selecting a matching target.

[0053] The classification unit (141) classifies a plurality of face template information stored in the storage unit (130) into a plurality of groups according to facial features. The similarity calculation unit (143) compares the face template information of the matching requesting user with the face template information of other users within the group to which the face template information of the matching requesting user belongs among the plurality of groups classified by the classification unit (141), and calculates the similarity.

[0054] That is, the classification unit (141) classifies a plurality of face template information stored in the storage unit (130) into a plurality of groups according to facial features, and the similarity calculation unit (143) compares the face template information only within the same group to which the matching request user belongs among the plurality of groups classified in this way, thereby eliminating the need to compare all of the data and completing the similarity calculation process more quickly.

[0055] This classification process can be implemented using the K-means clustering technique. K-means clustering is an unsupervised learning algorithm that divides data into multiple groups (clusters). This algorithm divides data based on the distance to the centroid of each cluster and, by calculating the distance between facial feature vectors, groups users with similar faces into the same cluster. The clustering process first randomly sets K cluster centers. The distance between each user's face template (facial feature vector) and each cluster center is calculated, and the user's face template is assigned to the closest cluster center. The cluster center is updated by calculating the average of all face vectors assigned to each cluster. During this process, the center of each cluster shifts according to the data distribution. This update process is repeated until the cluster centers no longer shift or a set number of iterations is reached. Once this process is complete, users with similar faces are grouped into the same cluster.

[0056] Users with similar facial features are pre-grouped into clusters, and when a matching request signal is sent, comparisons are performed only within those clusters. This reduces the matching time by performing matching on only a subset of the dataset, rather than the entire dataset.

[0057] Meanwhile, unlike this K-means clustering technique, the classification unit (141) can also classify face template information into multiple groups according to separate classification criteria.

[0058] These classification criteria can be transmitted from the user terminal (10) through the selection of the matching request user. For example, a screen for selecting facial features that are considered important among facial features can be displayed on the user terminal (10), and the user can be configured to select any one of the facial features displayed on the screen. The facial feature information to be selected can be, for example, face shape, eye shape, nose shape, mouth shape, etc. When the classification criteria according to these facial features are transmitted from the user terminal (10), the classification unit (141) classifies the entire facial template information into a plurality of groups according to the transmitted classification criteria.

[0059] For example, if a matching request user selects a face shape as a classification criterion, the classification unit (141) can classify the entire face template information into square, oval, round, etc. based on the face shape. The similarity calculation unit (143) can compare the face template information among the classified face shapes within the group (e.g., oval) to which the matching request user's face shape belongs to, and calculate the highest similarity.

[0060] At this time, the classification unit (141) may classify the data into multiple groups based on multiple classification criteria in advance and separately store the groups for each classification criterion in the storage unit (130). In this way, when the groups corresponding to each classification criterion are separately stored by the classification unit (141), after the matching request user selects a classification criterion, instead of performing the corresponding classification task, a specific group can be directly selected from among the already classified groups and similarity calculation can be performed within the selected group. Therefore, the similarity calculation task can be performed very quickly.

[0061] To summarize this process, as illustrated in FIG. 4, first, a matching request signal is transmitted from the user terminal (10) (S10), and classification criteria are selected and transmitted together with the matching request signal (S20). In the process of selecting the classification criteria, restriction conditions may also be selected and transmitted together. Thereafter, the matching server (100) determines whether restriction conditions have been transmitted from the user terminal (10) together with the matching request signal (S30), and selects users according to the restriction conditions through the primary selection unit (142) for the first time (S40). Thereafter, the similarity is calculated by comparing the facial template information of the users selected for the first time through the similarity calculation unit (143) (S50). In the process of calculating the similarity, the similarity may be calculated for only one of the groups classified according to the aforementioned classification criteria. Thereafter, the user with the highest similarity is selected as the matching target user (S60), and this is generated as matching information and provided to the user terminal (10) (S70).

[0062] In this way, the user who most resembles the matching request user can be selected and provided as matching information to the user.

[0063] Meanwhile, the matching target selection unit (144) may select the user with the highest similarity as the matching target user as described above, but alternatively, it may select multiple matching target users in order of high similarity up to a preset standard ranking (e.g., 5th place) so that the matching request user can select them, and select the user selected by the matching request user as the final matching target user to generate matching information.

[0064] At this time, the matching information generation unit may further include a selection analysis unit (145) that analyzes the result of the matching request user selecting the final matching target user and extracts the facial feature with the highest similarity among the facial features between the matching request user and the final matching target user.

[0065] The results analyzed by the selection analysis unit (145) can be reflected as feedback information in the future similarity calculation process. Specifically, the similarity calculation unit (143) can calculate similarity by assigning weights to the facial features extracted through the selection analysis unit (145).

[0066] For example, if a matching request user selects one of multiple users with a high similarity score as the final matching target user, the selection analysis unit (145) can compare the facial templates of the matching request user and the final matching target user to analyze which facial feature is most similar. For example, the similarity score of the eyes may be analyzed to be the highest at 90. In this case, since the matching request user perceives similar eyes as the most desirable, a weight may be given to the eye feature when calculating similarity in the future. In other words, if the eyes are similar, they can have a higher similarity score than other parts.

[0067] By analyzing the final match selection results for these users across multiple users, the similarity calculation method can be updated to apply weights to specific facial features not only for specific users but also for all users.

[0068] More accurate matching information can be generated through the process of calculating similarity using these weights.

[0069]

[0070] The above description is merely an illustrative illustration of the technical idea of ​​the present invention, and those skilled in the art will appreciate that various modifications and variations can be made without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate, rather than limit, the technical idea of ​​the present invention, and the scope of the technical idea of ​​the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of the rights of the present invention.

Claims

1. Includes a matching server that stores and analyzes a facial image transmitted from a user terminal to generate matching information and provides the generated matching information to the user terminal. The above matching server is An information collection unit that receives and collects user information and facial images from a user terminal; A feature extraction unit that extracts facial feature information using an artificial intelligence algorithm based on the facial image collected by the above information collection unit and creates a facial template; A storage unit that stores the above facial image and facial template information together with user information in a database; When a matching request signal is transmitted from a user terminal, a matching information generation unit that generates matching information by comparing the face template information of the user of the user terminal that transmitted the matching request signal with the face template information of another user stored in the storage unit; and An information providing unit that transmits the matching information generated by the above matching information generating unit to the user terminal that transmitted the above matching request signal. An artificial intelligence-based blind date matching system based on facial images, including:

2. In paragraph 1, The above matching information generation unit An artificial intelligence-based blind date matching system based on a facial image, which compares the facial template information of a matching request user, who is a user of a user terminal that has transmitted the above-mentioned matching request signal, with the facial template information of another user, extracts a user having the highest degree of similarity between the matching request user and the facial template information based on the comparison result, and matches the extracted user with the matching request user to create the above-mentioned matching information.

3. In paragraph 2, The above matching information generation unit A similarity calculation unit that calculates similarity by comparing the face template information of the matching request user stored in the storage unit with the face template information of another user; and A matching target selection unit that selects the user with the highest similarity as the matching target user based on the calculation result of the above similarity calculation unit and generates the matching information. An artificial intelligence-based blind date matching system based on a facial image, wherein the matching information includes a facial image and user information of the user to be matched, and a similarity value for the user to be matched.

4. In paragraph 3, The above matching information generation unit A primary selection unit that selects users according to the restriction conditions from among the user information stored in the storage unit, taking into account the restriction conditions transmitted from the user terminal. An artificial intelligence-based blind date matching system based on a facial image, wherein the similarity calculation unit further includes a similarity calculation unit for calculating a similarity for a user selected by the first selection unit.

5. In paragraph 4, The above restriction conditions are an artificial intelligence-based blind date matching system based on facial images, including gender conditions, age conditions, and region conditions.

6. In paragraph 3, The above matching information generation unit A classification unit that classifies a plurality of face template information stored in the above storage unit into multiple groups according to facial features. An artificial intelligence-based blind date matching system based on a facial image, wherein the similarity calculation unit calculates a similarity by comparing the facial template information of the matching request user with the facial template information of another user within a group to which the facial template information of the matching request user belongs among a plurality of groups classified by the classification unit.

7. In paragraph 6, The above classification unit classifies the facial features into multiple groups based on multiple classification criteria, and each classification criterion is classified into multiple groups. An artificial intelligence-based blind date matching system based on a facial image, wherein when a classification criterion is transmitted from a user terminal by an input of the matching requesting user, the similarity calculation unit calculates the similarity within a group to which the face template information of the matching requesting user belongs among a plurality of groups classified according to the transmitted classification criterion.

8. In paragraph 3, The above matching target selection section An artificial intelligence-based blind date matching system based on a facial image, which selects multiple matching target users in order of high similarity up to a preset standard ranking so that the matching request user can select them, and selects the user selected by the matching request user as the final matching target user and creates the matching information.

9. In paragraph 8, The above matching information generation unit A selection analysis unit that analyzes the result of the above matching request user selecting the final matching target user and extracts the facial feature with the highest similarity among the facial features between the above matching request user and the final matching target user. An artificial intelligence-based blind date matching system based on a facial image, wherein the similarity calculation unit calculates similarity by applying weights to facial features extracted through the selection analysis unit.

Images