TW201322015A - Community friend-making method using face recognition - Google Patents

Abstract

The present invention discloses a community friend-making method using face recognition, which includes: the entrant inputs the image information into the community friend-making platform including the facial image; setting the search criteria; conducting the face recognition for the facial image by referring to the pre-established face recognition database to generate the recognition result; and, completing the searching service according to the recognition result and the search criteria, which is to search for multiple similar members from the members of the community friend-making platform having the facial images similar to the facial image inputted by the entrant to generate the searching result of similar members, and the searching result of similar members may include the facial image, age, sex, occupation, place of residence or nationality for each similar member. The method according to the present invention can help the qualified entrants to conveniently find out the candidates from the existed members similar to a specific person as a reference for determining whether or not to proceed with the friend-making activities.

Description

Face recognition community dating method

The present invention relates to a community dating method, and more particularly to using a face recognition function to provide a search service required for community dating.

With the development of online information and the popularity of social networking platforms, users can easily join a specific social networking platform to obtain membership, and then can make friends with other members of the specific community dating platform.

Users typically choose to make friends with specific categories of members, such as having a specific skin tone, long image, or gender. Although the social dating platform usually provides photos and basic information of all members for reference, users need to view the images of the photos and the text content of the basic materials one by one, so that they can select the candidates that are needed in the desired way, so they not only consume Time, labor and inefficiency, and it is easy to ignore possible candidates due to visual fatigue, especially for well-known large community dating platforms with thousands, tens of thousands or even millions of members.

Because people's facial features, such as facial features, are the most representative of a person's appearance characteristics, but also the main target of visual identity, therefore, various face recognition techniques have been widely used in various identification fields, such as access control. Part or all replace the traditional guard function. However, the technique of face recognition has not been used in the general community dating platform.

Therefore, there is a need for a face recognition community dating method that can be used to assist the registrant in finding a suitable person to use as a basis for making friends, thereby solving the problem of the above-mentioned conventional techniques.

The main purpose of the present invention is to provide a face recognition community dating method, which is to provide a reference service for all qualified members of the social dating platform to provide face recognition by the face recognition, and the face recognition community dating method includes: The registrant inputs the image data to the social dating platform, and the image data includes at least a facial image; the registrant sets the search condition; the social friend platform refers to the pre-established face recognition database and inputs the facial image input to the registrant. Perform face recognition to generate identification results, and the face recognition database includes facial image data of registered qualified members; and search service based on identification results and search conditions, searching for members of the social dating platform Similar members who are similar to the facial images entered by the registrant to generate similar member search results, which may include facial images, age, gender, occupation, place of residence or nationality of each similar member. It may further include communication materials for each similar member, such as local phone number, mobile phone number, and e-mail. Address (e-mail address).

The face recognition community dating method of the present invention can assist a qualified registrant to conveniently find a plurality of candidates who are more similar to a specific person among the original members, thereby taking a reference as to whether or not to perform a friend activity.

Another object of the present invention is to provide a face recognition community dating method for identifying a registrant identity, and the face recognition community dating method includes: the registrant inputs image data to the social dating platform, and the The image data includes at least the face image of the registrant; the social dating platform refers to the face recognition database to perform face recognition on the registrant to generate the registrant identification result; and allows or denies the logged in user according to the entrant identification result. Community dating platform and display or transmit login identification results.

The above method can be used to prevent the registrant from impersonating other people's photos to scam other members, thereby improving the security of community friends. In addition, the present invention can also compare the photos uploaded by the members (the users of the dating platform basically upload a lot of photos) with the facial features captured and stored when the user registers and logs in, and the similarity is displayed next to those photos. Percentage, as a reference for other members to the authenticity of this person's photo.

The embodiments of the present invention will be described in more detail below with reference to the drawings and the <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt;

Referring to the first figure, a schematic diagram of the operation flow of the face recognition community dating method of the present invention. As shown in the first figure, the face recognition community dating method of the present invention includes steps S10, S20, S30, and S40 performed in sequence to provide a search service required for community dating by face recognition.

First, the face recognition community dating method of the present invention begins in step S10. The login user can use the client (client) or the server to log in to the social networking platform via the Internet, and then input or upload image data to the social dating platform. The client or the server may include a mobile phone, a computer or a terminal with internet access function, and the image data may include at least a facial image and/or a plurality of facial features, and the facial features may include a distance between the eyes, a skin color, and At least one of the aspect ratios of the face.

Next, proceeding to step S20, at least one search condition is set by the registrant, and the search condition may include at least one of the number of similar members after the search, gender, hobbies, occupation categories, nationality, place of residence, and age range.

Then, in step S30, the community dating platform can refer to the pre-established face recognition database to perform face recognition on the image data input or uploaded by the registrant, thereby generating identification results, wherein the face recognition database includes Face images or facial features of multiple qualified members.

Finally, in step S40, the community dating platform searches for at least one similar member of the community dating platform member similar to the face image input or uploaded by the registrant based on the identification result and the search condition, and generates each of the similar members. Similar member search results for similar members can be displayed or submitted to the registrant to provide a reference for making friends. In addition, similar member search results may further include at least one of facial images, age, gender, hobbies, occupations, places of residence, and nationality of each similar member, and may further include communication materials of each similar member, such as the city. Phone number, mobile number, email address.

For example, the registrant can upload photos of movie star Andy Lau and search for the top ten members with the highest similarity with Andy Lau as a reference for making friends or interactions, making it easier for registrants to view the members' images one by one. Find the right friends to choose.

In addition, the face recognition community dating method of the present invention can further perform the identity recognition process (not shown) of the registrant when the logged in user logs into the community dating platform, wherein the identity recognition process can include input or upload by the registrant first. The registrant image data is sent to the social dating platform, and the registrant image data includes at least the logged in person's face image and/or multiple facial features. Then, the social dating platform refers to the face recognition database to perform face recognition on the registrant to generate the registrant identification result. Finally, the social dating platform relies on the registrant identification result to allow or deny the logged in to the community dating platform, and can display or transmit the registrant identification result for the registrant to refer to.

The above-mentioned registrant identification result may include a percentage of facial similarity, wherein the facial similarity percentage refers to a facial image, a photo, and/or a plurality of facial features input by the registrant at the time of login, and the registrant's own registration. The percentage of similarity between stored facial features can be indicated next to the photo of the logged in person in the social dating platform and used as a reference for other members to verify the authenticity of the registrant's photo.

Using face recognition to determine the identity of the registrant can prevent fraudulent use of accounts or theft of photos, because many people are cheated on the dating site, and people who are interested in using beauty or handsome photos to pretend themselves, to scam .

The above is only a preferred embodiment for explaining the present invention, and is not intended to limit the present invention in any way, and any modifications or alterations to the present invention made in the spirit of the same invention. All should still be included in the scope of the intention of the present invention.

S10. . . Loginer enters image data

S20. . . Set search criteria

S30. . . Face recognition to generate identification results

S40. . . Search for multiple similar members to generate similar member search results

The first figure shows a schematic diagram of the operation flow of the method for making friends in the face recognition community of the present invention.

S10. . . Loginer enters image data

S20. . . Set search criteria

S30. . . Face recognition to generate identification results

S40. . . Search for multiple similar members to generate similar member search results

Claims (8)

A face recognition community dating method for searching a face recognition of a community dating platform, the face recognition community dating method includes: a login is accessed by a client or a server via a network After the community dating platform, input or upload an image data to the social networking platform, and the image data includes at least one facial image and/or multiple facial features; the registrant sets at least one search condition; the social friend The platform refers to a pre-established face recognition database to perform face recognition on the image data input or uploaded by the registrant, thereby generating a recognition result, and the face recognition database includes facial images of a plurality of qualified members or a facial feature; and the community dating platform searches for at least one similar member of the community dating platform that is similar to the facial image input or uploaded by the registrant based on the identification result and the at least one search condition And generate a similar member search result containing each similar member, and display or submit to the login to provide a reference for making friends. According to the face recognition community dating method described in claim 1, wherein the client or the server includes a mobile phone, a computer or a terminal with internet access function. The face recognition community dating method according to claim 1, wherein the facial features include at least one of a binocular distance, a skin color, and a face aspect ratio. The face recognition community dating method according to claim 1, wherein the search condition includes at least one of the number of similar members, gender, hobbies, occupation categories, nationality, place of residence, and age range. According to the face recognition community dating method described in claim 1, wherein the similar member search result further includes at least one of each similar member's facial image, age, gender, hobby, occupation, place of residence, and nationality. one. According to the face recognition community dating method described in claim 1, wherein the similar member search result further includes communication materials of each similar member, and the communication data includes a local phone number, a mobile phone number, and an email address. . According to the face recognition community dating method described in claim 1 of the patent application, the registrant's identity recognition process is further performed when the registrant logs into the community dating platform, and the identity recognition process includes: the registrant input or Uploading a registrant image data to the community dating platform, and the registrant image data includes at least the logged in person's face image and/or a plurality of facial features; the community dating platform refers to the face recognition database and The registrant performs face recognition to generate a registrant identification result; and based on the registrant identification result, the logged in user is allowed or denied to log in to the social networking platform, and the registrant identification result is displayed or transmitted. According to the face recognition community dating method described in claim 7, wherein the registrant identification result includes at least a face similarity percentage, and the facial similarity percentage is input by the registrant when logging in. The percentage of similarity between the face image, the photo and/or the face features, and the facial features stored by the registrant when registering, is used to mark the photo of the registrant in the social dating platform. As a reference for the authenticity of the photo of the registrant by other members.