Authentication method, authentication device and storage medium

Abstract

The present disclosure relates to an identity authentication method, an identity authentication device, and a storage medium. The identity authentication method is applied to a terminal, and the identity authentication method includes: in response to determining to authenticate a first user, displaying a face image for identity authentication and verification information for identity authentication, wherein the face image is a face image of a second user. Based on the verification information sent by the second user, the first user is authenticated. Through the identity authentication method provided by the present disclosure, the terminal can determine that the first user is a safe user based on the verification operation sent by the second user, thereby improving the security of the terminal's identity authentication.

Description

Technical Field

[0001] The present disclosure relates to the field of face recognition technology and terminal security, and in particular to an identity authentication method, an identity authentication device and a storage medium. Background Art

[0002] In the related technology, when the terminal performs identity authentication, it performs security verification based on relevant information such as ID card, bank card and mobile phone verification code. The use scenarios that require the terminal to perform identity authentication may include: account registration, password modification or payment, etc. However, when the terminal is lost or the identity information is leaked, others can use the terminal or the leaked identity information to obtain the verification code for identity authentication, which seriously affects the information security in the terminal. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides an identity authentication method, an identity authentication device and a storage medium.

[0004] According to a first aspect of an embodiment of the present disclosure, there is provided an identity authentication method, which is applied to a terminal, and the identity authentication method comprises: in response to determining to authenticate a first user, displaying a face image for the identity authentication and verification information for the identity authentication, wherein the face image is a face image of a second user, and authenticating the first user based on the verification information sent by the second user.

[0005] In one embodiment, the authenticating the first user includes: authenticating the first user based on a first account. The displaying of a facial image used for the authentication includes: sending a request to a cloud server, the request being used to request to obtain a facial image associated with the first account and used for the authentication of the first user. Receiving and displaying the facial image sent by the cloud server based on the request.

[0006] In another embodiment, the identity verification method further includes: displaying a control for changing a facial image. If the control is triggered, the facial image used for the identity verification is updated and displayed.

[0007] According to a second aspect of an embodiment of the present disclosure, another identity authentication method is provided, which is applied to a cloud server, and the identity authentication method includes: obtaining a request sent by a terminal, the request being used to request obtaining a facial image associated with a first account and used to authenticate the first user, the facial image being a facial image of a second user, and sending the facial image based on the request.

[0008] In one embodiment, the facial image for identity authentication of the first user is determined in the following manner: facial features are extracted from facial images stored in a facial image library associated with the first account, and multiple facial image clusters are obtained based on the extracted facial features. A first facial image cluster is determined among the multiple facial image clusters, and the first facial image cluster includes facial images of actual users using the first account. A first number of second facial image clusters whose intimacy with the actual user facial images is greater than an intimacy threshold are determined among the other facial image clusters, and the other facial image clusters are facial image clusters other than the first facial image cluster among the multiple facial image clusters. Based on the first number of second facial image clusters, a facial image for identity authentication is determined.

[0009] In another embodiment, determining the first facial image cluster from the plurality of facial image clusters includes: determining a first facial identification image based on the first account, the first facial identification image being used to uniquely identify an actual user using the first account. Determining a plurality of central facial feature vectors corresponding to the plurality of facial image clusters respectively, and determining a facial image cluster whose central facial feature vector and the facial feature vector of the first facial identification image have a similarity greater than a similarity threshold as the first facial image cluster. The central facial feature vector is the mean of the plurality of facial feature vectors included in the facial image cluster.

[0010] In another embodiment, determining a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold in other facial image clusters includes: respectively determining the number of facial images corresponding to each of the other facial image clusters. For each facial image cluster in the other facial image clusters, normalization processing is performed based on the number of facial images corresponding to each and the total number of the other facial image clusters to obtain a normalized value of each facial image cluster in the other facial image clusters, wherein a larger normalized value indicates that the facial image cluster is closer to the actual user facial image. In the order of the normalized values ​​from large to small, a first number of second facial image clusters are determined in the other facial image clusters.

[0011] In another embodiment, determining the facial image used for the identity authentication based on the first number of second facial image clusters includes: selecting central facial images corresponding to some second facial image clusters in the first number of second facial image clusters as the facial image used for the identity authentication.

[0012] According to a third aspect of an embodiment of the present disclosure, there is provided an identity authentication device, which is applied to a terminal, and comprises: a display unit, which is used to display a face image used for identity authentication and verification information used for identity authentication in response to determining to authenticate a first user, wherein the face image is a face image of a second user; and a verification unit, which is used to authenticate the first user based on the verification information sent by the second user.

[0013] In one embodiment, the identity authentication of the first user includes: identity authentication of the first user based on a first account. The display unit displays the facial image used for the identity authentication in the following manner: sending a request to a cloud server, the request being used to request to obtain a facial image associated with the first account and used for the identity authentication of the first user. Receiving and displaying the facial image sent by the cloud server based on the request.

[0014] In another embodiment, the display unit is further used to: display a control for changing the face image. If the control is triggered, the face image used for identity verification is updated and displayed.

[0015] According to a fourth aspect of an embodiment of the present disclosure, there is provided an identity authentication device, which is applied to a cloud server, and the identity authentication device includes: an acquisition unit, which is used to acquire a request sent by a terminal, the request being used to request to acquire a facial image associated with a first account and used to authenticate the first user, the facial image being a facial image of a second user; and a sending unit, which is used to send the facial image based on the request.

[0016] In one embodiment, the identity authentication device further includes: a determination unit. The determination unit determines a facial image for identity authentication of the first user in the following manner: facial features are extracted from facial images stored in a facial image library associated with the first account, and multiple facial image clusters are obtained based on the extracted facial features. Among the multiple facial image clusters, a first facial image cluster is determined, and the first facial image cluster includes facial images of actual users using the first account. Among other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial images is greater than an intimacy threshold are determined, and the other facial image clusters are facial image clusters other than the first facial image cluster in the multiple facial image clusters. Based on the first number of second facial image clusters, a facial image for identity authentication is determined.

[0017] In another embodiment, the determination unit determines the first facial image cluster from the plurality of facial image clusters in the following manner: based on the first account, determine a first facial identification image, the first facial identification image being used to uniquely identify an actual user using the first account. Determine the corresponding plurality of central facial feature vectors in the plurality of facial image clusters respectively, and determine the facial image cluster whose similarity between the central facial feature vector and the facial feature vector of the first facial identification image is greater than a similarity threshold as the first facial image cluster. The central facial feature vector is the mean of the plurality of facial feature vectors included in the facial image cluster.

[0018] In another embodiment, the determination unit determines a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold in other facial image clusters in the following manner: respectively determine the number of facial images corresponding to each of the other facial image clusters. For each facial image cluster in the other facial image clusters, normalization processing is performed based on the number of facial images corresponding to each and the total number of the other facial image clusters to obtain a normalized value of each facial image cluster in the other facial image clusters, wherein a larger normalized value indicates that the facial image cluster is closer to the actual user facial image. In the order of the normalized values ​​from large to small, a first number of second facial image clusters are determined in the other facial image clusters.

[0019] In another embodiment, the determination unit determines the facial image used for the identity authentication based on the first number of second facial image clusters in the following manner: selecting central facial images corresponding to some second facial image clusters in the first number of second facial image clusters as the facial image used for the identity authentication.

[0020] In yet another embodiment, the display unit is further configured to: display a control for changing a facial image, and if the control is triggered, update and display the facial image used for the identity verification.

[0021] According to a fifth aspect of an embodiment of the present disclosure, there is provided an identity authentication device, comprising: a memory for storing instructions; and a processor for calling the instructions stored in the memory to execute any one of the above-mentioned identity authentication methods.

[0022] According to a sixth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, in which instructions are stored. When the instructions are executed by a processor, any one of the above-mentioned identity authentication methods is executed.

[0023] The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: through the identity authentication method provided by the present disclosure, when verifying the identity of the first user, a facial image of the second user used to send verification information is provided to the first user, and then based on the verification information sent by the second user, the legitimacy of the identity of the first user can be determined, thereby improving the security of identity authentication of the terminal.

[0024] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0026] Figure 1 The present invention is a flowchart of an identity authentication method according to an exemplary embodiment.

[0027] Figure 2 The figure is a flow chart showing another identity authentication method according to an exemplary embodiment.

[0028] Figure 3 The figure is a flow chart of a method for determining a facial image according to an exemplary embodiment.

[0029] Figure 4 The figure is a flow chart of a method for extracting features from a facial image according to an exemplary embodiment.

[0030] Figure 5 The figure is a flow chart of a method for determining a facial image cluster according to an exemplary embodiment.

[0031] Figure 6 The figure is a flow chart of another method for determining facial image clusters according to an exemplary embodiment.

[0032] Figure 7 The present invention is a flowchart of confirming an association relationship according to an exemplary embodiment.

[0033] Figure 8 The invention is a block diagram of an identity authentication device according to an exemplary embodiment.

[0034] Fig. 9 The invention is a block diagram of another identity authentication device according to an exemplary embodiment.

[0035] Fig.10 The invention is a block diagram showing a device for identity authentication according to an exemplary embodiment.

[0036] Fig.11The invention is a block diagram showing a device for identity authentication according to an exemplary embodiment. DETAILED DESCRIPTION

[0037] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0038] In the related art, when a terminal performs identity authentication, it can perform authentication based on information such as an ID card, a bank card, and a mobile phone verification code. However, in most cases, the verification code used for identity authentication is sent to the terminal in the form of a text message, and then the current user of the terminal enters the verification code to perform identity authentication. However, when the terminal is lost or the identity information is leaked, others can use the terminal or the leaked identity information to obtain the verification code used for identity authentication, which seriously affects the information security in the terminal. What's worse, it is easy to cause damage to the property or even the reputation of the terminal holder.

[0039] In view of this, the present disclosure provides an identity authentication method, which provides a second user image for sending verification information to the first user when performing identity authentication for the first user. Based on the verification information sent by the second user, the legitimacy of the identity of the first user can be determined. The terminal can determine that the first user is a safe user based on the verification operation sent by the second user, thereby helping to improve the security of identity authentication of the terminal and reduce the possibility of information leakage in the terminal.

[0040] In the embodiment of the present disclosure, the first user can be understood as the current user of the terminal, or can also be understood as the user who triggers identity authentication. In the embodiment of the present disclosure, the second user can be understood as the user who sends the verification information provided by the terminal.

[0041] In one example, the type of terminal may include a mobile terminal, such as a mobile phone, a tablet, a notebook, etc. In another example, the structure of the terminal may include a double-sided screen terminal, a folding screen terminal, a full-screen terminal, etc.

[0042] Figure 1 is a flow chart of an identity authentication method according to an exemplary embodiment. Figure 1 As shown, the identity authentication method is used in a terminal, and includes the following steps S11 to S12.

[0043] In step S11 , in response to determining to authenticate the first user, a face image for identity authentication and authentication information for identity authentication are displayed.

[0044] In an embodiment of the present disclosure, when the first user triggers an operation for identity authentication, the terminal responds and displays a face image and verification information for identity authentication. The face image is a face image of a second user for the first user to select and determine the second user to whom verification information needs to be sent. The verification information is verification information sent by the first user to notify the second user. When the second user sends the verification information displayed by the terminal, it can be determined that the first user and the second user are acquainted. Thus, when identity authentication is performed, the legitimacy of the first user can be determined, reducing the risk that the first user is an illegal user. In one example, the first user notifies the second user of the verification information displayed by the terminal by an offline contact method, for example, by instant messaging methods such as calls, text messages, and emails. In another example, the terminal can provide a face image for identity authentication, and the second user is unique. In another example, the terminal can provide multiple face images for identity authentication, providing multiple possibilities for the first user to select the second user, so as to improve the verification pass rate. The verification information can be a verification code.

[0045] In step S12, the identity of the first user is authenticated based on the authentication information sent by the second user.

[0046] In the disclosed embodiment, if the second user sends verification information, it can be determined that the first user knows the second user corresponding to the face image used for identity authentication. Then, when the second user sends the verification information, it can be determined that the first user is a legitimate user and can be authenticated. This helps to improve the security of identity authentication at the terminal.

[0047] Through the above embodiment, based on the verification information displayed on the terminal sent by the second user, it can be determined that there is an interactive relationship between the first user and the second user used for identity authentication, and the verification information can be notified to the second user. Therefore, when identity authentication is performed for the first user, it can be helpful to improve the security and reliability of identity authentication, thereby reducing the risk that the first user is an illegal user and improving information security.

[0048] In one implementation scenario, the terminal responds according to the identity authentication triggered by the first user, and the terminal system automatically generates verification information. The first user identifies the second user through the face image for identity authentication displayed on the terminal, and notifies the second user of the verification information by offline contact. The second user receives the verification information sent by the first user, and the second user sends the verification information to the back-end server of the terminal. The back-end server verifies the legitimacy of the verification information based on the verification information sent by the second user. If the verification information is legitimate, the identity authentication of the first user is passed.

[0049] In another implementation scenario, if the first user is an illegal user, that is, a user who illegally holds the terminal. The terminal responds according to the identity authentication triggered by the first user, and the terminal system automatically generates verification information. Since the first user has nothing to do with the actual user of the terminal, when the first user performs identity authentication through the terminal, the first user cannot identify and contact the second user who needs to send verification information based on the face image displayed by the terminal for identity authentication. As a result, the second user cannot receive the verification information sent by the first user and cannot send the verification information to the back-end server of the terminal. Therefore, the back-end server cannot receive the verification information used to verify the legal identity of the first user. Therefore, the identity authentication of the first user will not pass.

[0050] In one embodiment, when the identity of the first user is authenticated, the identity of the first user can be authenticated based on the first account. The first account can be an account bound to the terminal, or it can also be understood as an account corresponding to an application that triggers the terminal to authenticate the identity. For example, the Xiaomi account of the terminal is used. Through the first account, the user information associated with the first account can be determined. Then, when the identity is authenticated, the legitimacy of the first user can be determined to prevent the terminal from being stolen by others.

[0051] In another embodiment, the facial image for identity authentication displayed in the terminal is obtained through a cloud server. When the first user triggers an identity authentication operation, the terminal sends a request to the cloud server to obtain a facial image associated with the first account and used to authenticate the first user. The cloud server is enabled to respond based on the request sent by the terminal, and sends the facial image associated with the first account and used to authenticate the first user to the terminal, so that the terminal can display according to the facial image sent by the cloud server. Acquiring the facial image for identity authentication through the cloud server helps save the storage space of the terminal and reduce the processing power consumption of the terminal.

[0052] In another embodiment, in order to avoid the facial image for identity authentication displayed by the terminal being an erroneous image, and to improve the verification pass rate and enhance the user experience, when the terminal displays the facial image for identity authentication and the verification information for identity authentication, a control for changing the facial image is also displayed. The terminal can respond according to the control being triggered, and update the displayed facial image for identity authentication, so that the first user can switch the facial image for identity authentication displayed by the terminal through the control, so that the first user can find a suitable second user for sending verification information and quickly complete the identity authentication.

[0053] Based on the same concept, the embodiment of the present disclosure also provides an identity authentication method applied to a cloud server.

[0054] Figure 2 FIG. 1 is a flow chart of a method for determining a face image according to an exemplary embodiment. Figure 2 As shown, the method for determining a face image includes the following steps S21 to S22.

[0055] In step S21, a request sent by the terminal is obtained.

[0056] In the disclosed embodiment, the request sent by the terminal is a request for obtaining a facial image associated with the first account and used to authenticate the first user. Through the first account, the user information associated with the first account can be determined, and then when the cloud server obtains the request, it can retrieve the facial image required for authenticating the first user from multiple cloud service images associated with the first account according to the first account in the request. Among them, the facial image is the facial image of the second user, which can be understood as the facial image of the user used to send the verification information provided by the terminal.

[0057] In step S22, the face image is sent based on the request.

[0058] In the disclosed embodiment, the retrieved facial image is sent to the terminal to complete a response to retrieving the facial image used to authenticate the first user.

[0059] Through the above-mentioned embodiments, the cloud server can retrieve the facial image used for identity authentication according to the request sent by the terminal, thereby helping to reduce the processing process of the terminal screening the facial image used for identity authentication from multiple cloud service images associated with the first account, helping to save the processing power consumption of the terminal and save the storage space of the terminal.

[0060] The embodiment of the present disclosure will now describe a process for determining a facial image for identity verification of a first user.

[0061] Figure 3is a flow chart of a method for determining a face image according to an exemplary embodiment. Figure 3 As shown, the method for determining a face image includes the following steps S31 to S34.

[0062] In step S31, facial features are extracted from facial images stored in a facial image library associated with the first account, and multiple facial image clusters are obtained by clustering based on the extracted facial features.

[0063] In an embodiment of the present disclosure, a cloud server may provide a cloud service, and cloud service images associated with multiple accounts may be stored in the cloud service, wherein the cloud service images include facial images. Different accounts correspond to different facial image libraries in the cloud server. In order to facilitate the determination of the facial image used for identity authentication in the first account, facial features are extracted from the facial images in the facial image library associated with the first account, and facial image features corresponding to each facial image are obtained. Based on a clustering algorithm, the facial image features corresponding to each extracted facial image are clustered to obtain multiple facial image clusters. Among them, the facial image of the same person corresponds to a facial image cluster. Thereby, the facial images in the facial image library associated with the first account are classified to determine the facial category involved in the first associated account. In one example, the clustering algorithm may include: a k-means clustering algorithm and a hierarchical clustering algorithm. Among them, the k-means clustering algorithm is an unsupervised algorithm with the characteristics of fast convergence speed, easy implementation, excellent clustering effect, strong interpretability of the algorithm, and the only parameter that needs to be adjusted is the number of clusters k. The hierarchical clustering algorithm is a clustering algorithm that calculates the similarity between nodes through a certain similarity measure, sorts them from high to low by similarity, and gradually reconnects the nodes.

[0064] In step S32, a first facial image cluster is determined among a plurality of facial image clusters.

[0065] In the disclosed embodiment, in order to facilitate the determination of the intimacy relationship between the faces of each facial image cluster and the actual user of the first account, the first facial image cluster is first determined among multiple facial image clusters. Among them, the first facial image cluster can be characterized as facial images of actual users using the first account. It can be further determined that among the multiple facial image clusters, the facial images corresponding to other facial image clusters except the first facial image cluster are all faces that have an associated relationship with the actual user of the first account. Therefore, when determining the facial image for identity authentication, it can be determined based on the interpersonal network of the actual user of the first account, so that the facial image displayed by the terminal for identity authentication is more reasonable.

[0066] In step S33, a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold are determined from other facial image clusters.

[0067] In the disclosed embodiment, other facial image clusters are facial image clusters other than the first facial image cluster in multiple facial image clusters, that is, facial images corresponding to other facial image clusters are facial images other than facial images of actual users of the first account. The existence of other facial image clusters indicates that the faces corresponding to other facial image clusters and the actual users of the first account have intersections in life. In order to improve the pass rate of identity authentication and improve the security of identity authentication, the association relationship between other facial image clusters and actual user facial images can be determined based on the intimacy between other facial image clusters and actual user facial images. The higher the intimacy, the more intersections that represent the relationship with the actual user facial image, and the closer the relationship.

[0068] By setting the intimacy threshold, the face images of other face image clusters and the face images of the actual user that are in a close relationship can be screened out. If there is a second face image cluster in the other face image clusters whose intimacy with the face image of the actual user is greater than the intimacy threshold, it indicates that the face corresponding to the second face image cluster and the actual user are in a close relationship with too many intersections. If there is a third face image cluster in the other face image clusters whose intimacy with the face image of the actual user is equal to or equal to the intimacy threshold, it indicates that the face corresponding to the third face image cluster and the actual user are in a casual relationship or a distant relationship with too few intersections. Then, according to the intimacy between each face image cluster in the other face image clusters and the face image of the actual user, the first number of second face image clusters whose intimacy with the face image of the actual user is greater than the intimacy threshold are determined. So that when the first user is a legitimate user, the second user for identity authentication determined in the first number of second face image clusters can help the first user to successfully complete identity authentication. Then it helps to promote the process of identity authentication and improve the pass rate of identity authentication.

[0069] In step S34, facial images for identity authentication are determined based on the first number of second facial image clusters.

[0070] In the example of the present disclosure, the first number can be any natural number, and the faces corresponding to the first number of second face image clusters belong to one or more faces that have a close relationship with the actual user of the first account. Then, when determining the face image for identity authentication, one or more face images corresponding to the second face image clusters can be selected from the first number of second face image clusters as the face image for identity authentication. In one example, if the first number is small, the face images corresponding to the first number of second face image clusters are the face images for identity authentication. In another example, if the first number is large, the face images for identity authentication can be determined by a specified determination method. The specified determination method can include: random selection or selection based on the size of intimacy. In another example, when determining the face image for identity authentication based on the size of intimacy, the first number of second face image clusters can be sorted from large to small according to the intimacy between the actual user of the first account, and based on the order of precedence, a specified number of second face image clusters are selected from the first number as the face image clusters for identity authentication, and then the face images corresponding to the specified number of second face image clusters are the face images for identity authentication. Determining the facial image used for identity authentication based on a specified number of second facial image clusters helps ensure that the facial image used for identity authentication has a high degree of intimacy with the actual user of the first account. Furthermore, when performing identity authentication, the second user corresponding to the facial image used for identity authentication can effectively help the terminal determine the legitimacy of the first user, thereby improving the security of information protection.

[0071] Through the above embodiment, the facial image used for identity authentication is determined based on the facial image in the facial image library associated with the first account, which can make full use of the interpersonal relationship of the actual user of the first account, so that the facial image determined for identity authentication is a user who has a close relationship with the actual user of the first account. Therefore, when performing identity authentication, the second user can help the terminal determine that the first user currently performing identity authentication is a legitimate user, so as to ensure the security of information and avoid information leakage of the actual user of the first account.

[0072] In one embodiment, extracting facial image features from the facial image library associated with the first account may be performed in the following manner.

[0073] Figure 4 FIG. 1 is a flow chart of a method for extracting features from a facial image according to an exemplary embodiment. Figure 4 As shown, the method for extracting features from facial images includes the following steps S41 to S43.

[0074] In step S41, face position information of a face image in a face image library associated with the first account is obtained through face detection.

[0075] In the disclosed embodiment, face detection is performed on the face image in the face image library associated with the first account, and the face position in the face image is determined by the marked face frame, thereby obtaining the face position information in the face image. Wherein, the face position information can be represented by [x, y, w, h]. x represents the horizontal coordinate of the point in the upper left corner of the face frame on the face image; y represents the vertical coordinate of the point in the upper left corner of the face frame on the face image; w represents the width of the face frame; h represents the height of the face frame. In one example, the face detection algorithm can be detected using a traditional algorithm. The traditional algorithm may include: a face image detection algorithm formed by combining the Haar algorithm and the Adaboost algorithm. Wherein, the Haar algorithm is a rectangular feature algorithm, and the Adaboost algorithm is an algorithm based on a cascade classification model. Or a face target detection algorithm formed by combining the Histogram of Oriented Gradient (HOG) algorithm and the Support Vector Machine (SVM) algorithm. In another example, the face detection algorithm can be detected using a deep learning algorithm. Deep learning algorithms may include: Region-Convolutional Neural Networks (R-CNN) series algorithms, SSD (Single Shot MultiBox Detector) series algorithms, or YOLO (You Only Look Once) series algorithms. Among them, SSD is a multi-target detection algorithm that directly predicts the target category and performs multi-box prediction. YOLO is an object recognition and positioning algorithm based on a deep neural network.

[0076] In step S42, based on the face position information, face key points are detected on the face image, and the face image is corrected to obtain a corrected face image.

[0077] In the disclosed embodiment, a face image with face position information is input into a key point detection model to obtain face key points corresponding to the face image. Among them, the face key points may include left eye key points, right eye key points, nose tip key points, left corner of mouth key points and right corner of mouth key points in the face image. The key point detection model may use a deep learning algorithm to perform key point detection. Based on the obtained face key points, a transformation matrix corresponding to the face image is obtained. Then, based on the transformation matrix, the face image is corrected through affine transformation to obtain a corrected face image.

[0078] In step S43, facial features are extracted from the corrected facial image to obtain facial image features in the facial image library associated with the first account.

[0079] In one example, the vector dimension of the facial image feature may be 128, 256, 512 or 1024. Different loss functions in the algorithm model will result in different vector dimensions of the facial image feature.

[0080] In one embodiment, when determining the first face image cluster from the plurality of face image clusters, the first face identification image that identifies the actual user of the first account may be used to determine the first face image cluster. The determination process may be as follows: Figure 5 As shown, Figure 5 The figure is a flow chart of a method for determining a facial image cluster according to an exemplary embodiment.

[0081] In step S51, a first facial identification image is determined based on the first account.

[0082] In the disclosed embodiment, the first facial identification image is an image used to uniquely identify the actual user who uses the first account. For example, the facial identification image may be the facial image on the ID card of the actual user of the first account after real-name authentication. The first account is associated with the first facial identification image. The first facial identification image can be determined through the first account. Furthermore, based on the first facial identification image, the facial image features of the actual user of the first account can be clarified, so that the first facial image cluster with the same facial features as the actual user of the first account can be quickly determined in multiple facial image clusters based on facial image feature comparison.

[0083] In step S52, multiple central face feature vectors corresponding to multiple face image clusters are respectively determined, and a face image cluster whose similarity between the central face feature vector and the face feature vector of the first face identification image is greater than a similarity threshold is determined as the first face image cluster.

[0084] In the disclosed embodiment, the central face feature vectors in the plurality of face image clusters are determined respectively. In one example, the central face feature vector can be represented as the mean of the plurality of face feature vectors included in the face images in the corresponding face image cluster. For example, taking a single face image cluster as an example, all the face feature vectors in the face image cluster are recorded as {X i1 ,X i2 ,X i3 ,…,X iN}, where i represents a face and N represents the feature vector dimension. 11The meaning of can be understood as the first-dimensional eigenvalue corresponding to the first face image in the face image cluster. Since the face feature vectors of each face image are extracted using the same algorithm, the dimensions of the face feature vectors corresponding to each face image are the same. Then the central face feature vector is obtained using the following formula: 1 / i*∑{x i1 ,x i2 ,x i3 ,…,x iN}, and obtain the mean of the facial feature vectors of the facial image cluster, that is, obtain the central facial feature vector. In another example, the central facial feature vector can be represented as the feature vector corresponding to the most representative facial image in the facial image cluster to which it corresponds.

[0085] The obtained central face feature vector is compared with the face feature vector of the first face identification image for similarity. The higher the similarity, the closer the similarity between the two. In one example, the similarity between the two can be measured by cosine similarity. The cosine calculation formula of the two vectors is: The value range of cos(θ) is between [-1,1]. In order to make it easier to clarify the similarity between the central face feature vector and the face feature vector of the first face identification image, cos(θ) is normalized using the formula (cos(θ)+1)*0.5 so that the value range of cos(θ) is between [0,1]. When comparing the similarity between the central face feature vector and the face feature vector of the first face identification image, X i or Y i represents the central face feature vector, Y i or X i A facial feature vector representing the first face identification image. A similarity threshold is preset, and the value range of the similarity threshold is between [0,1]. If the obtained similarity is greater than the similarity threshold, the similarity between the central face feature vector and the facial feature vector of the first face identification image is extremely high, and they can be approximated as facial feature vectors of the same face. If the obtained similarity is less than or equal to the similarity threshold, the similarity between the central face feature vector and the facial feature vector of the first face identification image is low, and the two are facial feature vectors of different faces. Among them, the similarity threshold can be a threshold set artificially based on experience, for example: the similarity threshold is 0.8. Then, the multiple central face feature vectors corresponding to the multiple face image clusters are respectively judged for similarity with the facial feature vector of the first face identification image, and the face image cluster whose similarity between the central face feature vector and the facial feature vector of the first face identification image is greater than the similarity threshold is determined as the first face image cluster.

[0086] In another embodiment, the process of determining the first number of second facial image clusters may be determined in the following manner.

[0087] Figure 6 FIG. 1 is a flow chart of a method for determining a face image cluster according to an exemplary embodiment. Figure 6 As shown, the method for determining a face image cluster includes the following steps S61 to S63.

[0088] In step S61, the number of facial images corresponding to other facial image clusters is determined respectively.

[0089] In the disclosed embodiment, the number of facial images corresponding to each of the other facial image clusters can reflect the intimacy between the faces corresponding to each of the other facial image clusters and the actual user of the first account. The more facial images there are in a facial image cluster, the more frequent the interaction between the actual user of the first account and the face corresponding to the facial image cluster, and the closer the relationship between the two. In order to determine the intimacy between the other facial image clusters and the actual user of the first account, the number of facial images corresponding to each of the other facial image clusters is first determined, so as to determine the intimacy between each face in the other facial image clusters and the actual user of the first account based on statistics.

[0090] In step S62, normalization processing is performed on each face image cluster in other face image clusters based on the number of face images corresponding to each face image cluster and the total number of other face image clusters to obtain normalized values ​​of each face image cluster in other face image clusters.

[0091] In the disclosed embodiment, since the number of face images involved in each face image cluster in other face image clusters is different, in order to make the obtained intimacy more reasonable and objective, the number of face images corresponding to each face image cluster in other face image clusters and the total number of other face image clusters are normalized so that the obtained normalized value belongs to [0,1]. Then, through the normalized value corresponding to each face image cluster in other face image clusters, the intimacy between each face image cluster and the actual user of the first account can be intuitively and quickly distinguished. Among them, the larger the normalized value, the closer the intimacy between the face image cluster and the actual user's face image. The normalization formula can be: S i =e i / ∑ j e j , where i represents the number of face images corresponding to a specific face image cluster in other face image clusters, and j represents the total number of other face image clusters.

[0092] In one implementation scenario, the total number of other face image clusters j=10, and the number of face images i corresponding to each face image cluster in the other face image clusters is 100, 10, 20, 40, 50, 126, 320, 51, 69, 78, respectively, which can be represented by a set, that is, S={100, 10, 20, 40, 50, 126, 320, 51, 69, 78}. Then, when normalizing, any number of face images in S={100, 10, 20, 40, 50, 126, 320, 51, 69, 78} and 10 are substituted into the above normalization formula. Thus, the normalized values ​​corresponding to each face image cluster in the other face image clusters are obtained.

[0093] In step S63, a first number of second face image clusters are determined from other face image clusters in descending order of normalized values.

[0094] In the disclosed embodiment, the larger the normalized value, the closer the intimacy between the face image cluster and the face image of the actual user. In order to ensure the information security of the actual user of the first account, ensure the smooth progress of identity authentication, and improve the identity authentication pass rate, the face image clusters in other face image clusters are sorted from small to large according to their corresponding normalized values, and the face image clusters of the first number with higher intimacy are selected as the second face image clusters. The first number can be one or more.

[0095] In one embodiment, when the first number is large, part of the second facial image cluster can be selected as the facial image displayed for identity authentication. And in order to facilitate the first user to clearly distinguish the second user corresponding to the displayed facial image, when selecting the facial image for display, the central facial image corresponding to the second facial image cluster is selected as the facial image for identity authentication. In one example, the central facial image can be characterized as the image that best represents the facial image in its corresponding second facial image cluster, such as a frontal image of the corresponding face of the facial image. In another example, the central facial image can be characterized as a facial image obtained based on the central facial feature vector corresponding to its corresponding second facial image cluster.

[0096] In another embodiment, multiple other accounts corresponding to each face image cluster in other face image clusters are determined, and then multiple face images associated with other accounts are obtained through multiple other accounts, and then multiple face image clusters corresponding to multiple other accounts are obtained. Then, when determining the close relationship between each face image cluster in other face image clusters and the actual user of the first account, the association relationship between the actual user of the first account and the users of multiple other accounts corresponding to other face image clusters can also be determined. Thereby, the interaction between different accounts is realized and the connection between different users is established. Based on the connection between the users of other accounts and the actual user of the first account, the mutual recognition between the users corresponding to other accounts and the actual user of the first account can be determined, and then when determining the close relationship between each face image cluster in other face image clusters and the actual user of the first account, combined with the connection between the users corresponding to other accounts and the actual user of the first account, the conclusion of the close relationship can be made more convincing. For example: A is the actual user of the first account, and each face image cluster in other face image clusters corresponds to the face images of users of three accounts B, C, and D. When determining the intimacy relationship between A and B, C, and D, in the face image libraries associated with the three accounts B, C, and D, based on cosine similarity, determine whether there is a face image cluster corresponding to A in B, whether there is a face image cluster corresponding to A in C, and whether there is a face image cluster corresponding to A in D. When there is a face image cluster whose similarity with the face image cluster corresponding to A is greater than the similarity threshold in the face image library associated with any one or more accounts of B, C, and D, it can be determined that A has an interactive relationship with it and the two users know each other. In this way, the accounts of two users can be associated together, for example, by mobile phone numbers. The value range of the similarity threshold is between [0,1]. In one example, the similarity threshold can be a threshold set manually based on experience. For example: the similarity threshold value is 0.8.

[0097] In one implementation scenario, when determining the association relationship between user A and user B, the confirmation process can be as follows: Figure 7 As shown. Among them, Figure 7It is a flowchart for confirming the association relationship. Face clustering service is performed for the cloud service pictures stored in the cloud server of user A and the cloud service pictures stored in the cloud server of user B, that is, face feature extraction is performed on the cloud service pictures of user A, and multiple face image clusters corresponding to user A are obtained based on the extracted face feature clustering, and face feature extraction is performed on the cloud service pictures of user B, and multiple face image clusters corresponding to user B are obtained based on the extracted face feature clustering. The cloud service pictures of user A and the cloud service pictures of user B include face images. Among them, cluster 11 represents the first dimension feature value of the first face image in the first face image cluster of user A. Cluster 12 represents the second dimension feature value of the first face image in the first face image cluster of user A. Cluster 1N represents the Nth dimension feature value of the first face image in the first face image cluster of user A. Cluster 21 represents the first dimension feature value of the first face image in the first face image cluster of user B. Cluster22 represents the second dimensional feature value of the first face image in the first face image cluster of user B. Cluster2N represents the Nth dimensional feature value of the first face image in the first face image cluster of user B. Then, among the multiple face image clusters corresponding to user A, the first face image cluster of user A is determined; and among the multiple face image clusters corresponding to user B, the first face image cluster of user B is determined. Based on the similarity calculation, if the multiple face image clusters corresponding to user A include some face image clusters of user B, and the multiple face image clusters corresponding to user B include some face image clusters of user A, then there can be an association relationship between user A and user B, and the two can be associated.

[0098] Based on the same concept, an embodiment of the present disclosure also provides an identity authentication device applied to a terminal.

[0099] It is understandable that the identity authentication device provided by the embodiment of the present disclosure includes hardware structures and / or software modules corresponding to the execution of each function in order to realize the above functions. In combination with the units and algorithm steps of each example disclosed in the embodiment of the present disclosure, the embodiment of the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the technical solution of the embodiment of the present disclosure.

[0100] Figure 8 is a block diagram of an identity authentication device according to an exemplary embodiment. Figure 8The identity authentication device 100 includes a display unit 101 and a verification unit 102.

[0101] The display unit 101 is used for displaying a face image for identity authentication and verification information for identity authentication in response to determining to authenticate the first user, where the face image is a face image of the second user.

[0102] The verification unit 102 is used to perform identity verification on the first user based on the verification information sent by the second user.

[0103] In one embodiment, authenticating the first user includes: authenticating the first user based on the first account. The display unit 101 displays the face image for identity authentication in the following manner: sending a request to the cloud server, the request being used to request to obtain the face image associated with the first account and used for identity authentication of the first user. Receive and display the face image sent by the cloud server based on the request.

[0104] In another embodiment, the display unit 101 is further used to: display a control for changing the face image. If the control is triggered, the face image for identity verification is updated and displayed.

[0105] Based on the same concept, the embodiment of the present disclosure also provides another identity authentication device applied to a cloud server.

[0106] Fig. 9 is a block diagram of an identity authentication device according to an exemplary embodiment. Fig. 9 The identity authentication device 200 includes an acquisition unit 201 and a sending unit 202.

[0107] The acquisition unit 201 is used to acquire a request sent by a terminal, where the request is used to request to acquire a facial image associated with a first account and used to authenticate the identity of the first user, where the facial image is a facial image of a second user.

[0108] The sending unit 202 is used to send the face image based on the request.

[0109] In one embodiment, the identity authentication device 200 also includes: a determination unit 203. The determination unit 203 determines the facial image for identity authentication of the first user in the following manner: facial features are extracted from facial images stored in a facial image library associated with the first account, and multiple facial image clusters are obtained based on the extracted facial features. Among the multiple facial image clusters, a first facial image cluster is determined, and the first facial image cluster includes facial images of actual users using the first account. Among the other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial images is greater than an intimacy threshold are determined, and the other facial image clusters are facial image clusters other than the first facial image cluster in the multiple facial image clusters. Based on the first number of second facial image clusters, a facial image for identity authentication is determined.

[0110] In another embodiment, the determination unit 203 determines the first face image cluster from the multiple face image clusters in the following manner: based on the first account, determine the first face identification image, the first face identification image is used to uniquely identify the actual user who uses the first account. Determine the central face feature vectors in the multiple face image clusters respectively, and determine the face image clusters whose similarity between the central face feature vector and the face feature vector of the first face identification image is greater than the similarity threshold as the first face image cluster.

[0111] In another embodiment, the determination unit 203 determines the first number of second facial image clusters whose intimacy with the actual user's facial image is greater than the intimacy threshold in other facial image clusters in the following manner: respectively determine the number of facial images corresponding to each of the other facial image clusters. For each facial image cluster in the other facial image clusters, normalization processing is performed based on the number of facial images corresponding to each and the total number of other facial image clusters to obtain a normalized value of each facial image cluster in the other facial image clusters, wherein a larger normalized value indicates that the facial image cluster is closer to the actual user's facial image. In the order of the normalized values ​​from large to small, the first number of second facial image clusters are determined in the other facial image clusters.

[0112] In another embodiment, the determination unit 203 determines the facial image for identity authentication based on the first number of second facial image clusters in the following manner: central facial images corresponding to some of the second facial image clusters in the first number of second facial image clusters are selected as the facial images for identity authentication.

[0113] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0114] Fig.10 is a block diagram of a device 300 for identity authentication according to an exemplary embodiment. For example, the device 300 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0115] Reference Fig.10 , the device 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 312 , a sensor component 314 , and a communication component 316 .

[0116] The processing component 302 generally controls the overall operation of the device 300, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 302 may include one or more modules to facilitate the interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate the interaction between the multimedia component 308 and the processing component 302.

[0117] The memory 304 is configured to store various types of data to support operations on the device 300. Examples of such data include instructions for any application or method operating on the device 300, contact data, phone book data, messages, pictures, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0118] The power component 306 provides power to the various components of the device 300. The power component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 300.

[0119] The multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 includes a front camera and / or a rear camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0120] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC), and when the device 300 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 304 or sent via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.

[0121] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0122] The sensor assembly 314 includes one or more sensors for providing various aspects of the status assessment of the device 300. For example, the sensor assembly 314 can detect the open / closed state of the device 300, the relative positioning of components, such as the display and keypad of the device 300, the sensor assembly 314 can also detect the position change of the device 300 or a component of the device 300, the presence or absence of user contact with the device 300, the orientation or acceleration / deceleration of the device 300, and the temperature change of the device 300. The sensor assembly 314 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 314 can also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 can also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0123] The communication component 316 is configured to facilitate wired or wireless communication between the device 300 and other devices. The device 300 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0124] In an exemplary embodiment, the device 300 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform any of the above-mentioned identity authentication methods.

[0125] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, and the instructions can be executed by the processor 320 of the device 300 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0126] Fig.11 4 is a block diagram of a device 400 for identity authentication according to an exemplary embodiment. For example, the device 400 may be provided as a server. Fig.11 , the apparatus 400 includes a processing component 422, which further includes one or more processors, and a memory resource represented by a memory 432 for storing instructions executable by the processing component 422, such as an application. The application stored in the memory 432 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 422 is configured to execute instructions to perform any of the above-mentioned authentication methods.

[0127] The device 400 may also include a power supply component 426 configured to perform power management of the device 400, a wired or wireless network interface 450 configured to connect the device 400 to a network, and an input / output (I / O) interface 458. The device 400 may operate based on an operating system stored in the memory 432, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0128] It is further understood that in the present disclosure, "plurality" refers to two or more than two, and other quantifiers are similar thereto. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. The singular forms "a", "the" and "the" are also intended to include plural forms, unless the context clearly indicates other meanings.

[0129] It is further understood that the terms "first", "second", etc. are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other, and do not indicate a specific order or degree of importance. In fact, the expressions "first", "second", etc. can be used interchangeably. For example, without departing from the scope of the present disclosure, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information.

[0130] It can be further understood that, unless otherwise specified, “connection” includes a direct connection without other components between the two, and also includes an indirect connection with other components between the two.

[0131] It is further understood that, although the operations are described in a specific order in the drawings in the embodiments of the present disclosure, it should not be understood as requiring the operations to be performed in the specific order shown or in a serial order, or requiring the execution of all the operations shown to obtain the desired results. In certain environments, multitasking and parallel processing may be advantageous.

[0132] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0133] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An identity authentication method, characterized in that: Applied to a terminal, the identity authentication method includes: In response to determining to authenticate the first user, displaying a facial image used for the authentication and verification information used for the authentication, wherein the facial image is a facial image of the second user; authenticating the first user based on the verification information sent by the second user; The displaying of the facial image used for identity verification includes: Sending a request to a cloud server, the request being used to request obtaining a facial image associated with a first account and used for performing the identity authentication of the first user; receiving and displaying the face image sent by the cloud server based on the request; The facial image for the identity verification of the first user is determined in the following manner: Extracting facial features from facial images stored in a facial image library associated with the first account, and clustering the extracted facial features to obtain a plurality of facial image clusters; Determine a first facial image cluster among the plurality of facial image clusters, wherein the first facial image cluster includes facial images of actual users using the first account; Determine, from other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold, wherein the other facial image clusters are facial image clusters other than the first facial image cluster among the multiple facial image clusters; Based on the first number of second facial image clusters, a facial image for performing the identity authentication is determined.

2. The identity authentication method according to claim 1, characterized in that: The authenticating the first user includes: authenticating the first user based on a first account.

3. The identity authentication method according to claim 1 or 2, characterized in that: The identity authentication method further includes: Displays controls for replacing a face image; If the control is triggered, the face image used for the identity authentication is updated and displayed.

4. An identity authentication method, characterized in that: Applied to a cloud server, the identity authentication method includes: Obtaining a request sent by the terminal, the request being used to request obtaining a facial image associated with the first account and used to authenticate the first user, the facial image being a facial image of the second user; Sending the facial image based on the request; The facial image for identity authentication of the first user is determined in the following manner: Extracting facial features from facial images stored in a facial image library associated with the first account, and clustering the extracted facial features to obtain a plurality of facial image clusters; Determine a first facial image cluster among the plurality of facial image clusters, wherein the first facial image cluster includes facial images of actual users using the first account; Determine, from other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold, wherein the other facial image clusters are facial image clusters other than the first facial image cluster among the multiple facial image clusters; Based on the first number of second facial image clusters, a facial image for performing the identity authentication is determined.

5. The identity authentication method according to claim 4, characterized in that: The step of determining a first face image cluster among the plurality of face image clusters comprises: Based on the first account, determine a first face identification image, where the first face identification image is used to uniquely identify an actual user who uses the first account; Respectively determine a plurality of central face feature vectors corresponding to the plurality of face image clusters, and determine a face image cluster whose similarity between the central face feature vector and the face feature vector of the first face identification image is greater than a similarity threshold as a first face image cluster; The central facial feature vector is the average of multiple facial feature vectors included in the facial image cluster.

6. The identity authentication method according to claim 4 or 5, characterized in that: The step of determining, from other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold, comprises: respectively determining the number of face images corresponding to each of the other face image clusters; For each face image cluster in the other face image clusters, normalization processing is performed based on the number of face images corresponding to each face image cluster and the total number of the other face image clusters to obtain a normalized value of each face image cluster in the other face image clusters, wherein a larger normalized value indicates that the face image cluster is closer to the actual user face image; A first number of second human face image clusters are determined from the other human face image clusters in descending order of the normalized values.

7. The identity authentication method according to claim 4, characterized in that: The determining of the facial image used for the identity authentication based on the first number of second facial image clusters includes: Central facial images corresponding to some second facial image clusters are selected from the first number of second facial image clusters as facial images for performing the identity authentication.

8. An identity verification device, characterized in that: Applied to a terminal, the identity verification device comprises: A display unit, configured to display a facial image used for performing identity authentication and verification information used for performing identity authentication in response to determining to perform identity authentication on the first user, wherein the facial image is a facial image of the second user; a verification unit, configured to authenticate the first user based on the verification information sent by the second user; The display unit displays the face image used for identity verification in the following manner: Sending a request to a cloud server, the request being used to request obtaining a facial image associated with a first account and used for performing the identity authentication of the first user; receiving and displaying the face image sent by the cloud server based on the request; The facial image for the identity verification of the first user is determined in the following manner: Extracting facial features from facial images stored in a facial image library associated with the first account, and clustering the extracted facial features to obtain a plurality of facial image clusters; Determine a first facial image cluster among the plurality of facial image clusters, wherein the first facial image cluster includes facial images of actual users using the first account; Determine, from other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold, wherein the other facial image clusters are facial image clusters other than the first facial image cluster among the multiple facial image clusters; Based on the first number of second facial image clusters, a facial image for performing the identity authentication is determined.

9. The identity verification device according to claim 8, characterized in that: The authenticating the first user includes: authenticating the first user based on a first account.

10. The identity verification device according to claim 8 or 9, characterized in that: The display unit is also used for: Displays controls for replacing a face image; If the control is triggered, the face image used for the identity authentication is updated and displayed.

11. An identity verification device, characterized in that: Applied to a cloud server, the identity verification device comprises: an acquiring unit, configured to acquire a request sent by the terminal, wherein the request is used to request to acquire a facial image associated with the first account and used to authenticate the first user, wherein the facial image is a facial image of the second user; A sending unit, configured to send the facial image based on the request; The identity authentication device further includes: a determination unit; the determination unit determines the face image for identity authentication of the first user in the following manner: Extracting facial features from facial images stored in a facial image library associated with the first account, and clustering the extracted facial features to obtain a plurality of facial image clusters; Determine a first facial image cluster among the plurality of facial image clusters, wherein the first facial image cluster includes facial images of actual users using the first account; Determine, from other facial image clusters, a first number of second facial image clusters whose intimacy with the actual user facial image is greater than an intimacy threshold, wherein the other facial image clusters are facial image clusters other than the first facial image cluster among the multiple facial image clusters; Based on the first number of second facial image clusters, a facial image for performing the identity authentication is determined.

12. The identity verification device according to claim 11, characterized in that: The determining unit determines the first face image cluster from the plurality of face image clusters in the following manner: Based on the first account, determine a first face identification image, where the first face identification image is used to uniquely identify an actual user who uses the first account; Respectively determine a plurality of central face feature vectors corresponding to the plurality of face image clusters, and determine a face image cluster whose similarity between the central face feature vector and the face feature vector of the first face identification image is greater than a similarity threshold as a first face image cluster; The central facial feature vector is the average of multiple facial feature vectors included in the facial image cluster.

13. The identity verification device according to claim 11 or 12, characterized in that: The determining unit determines a first number of second face image clusters whose intimacy with the actual user face image is greater than an intimacy threshold from other face image clusters in the following manner: respectively determining the number of face images corresponding to each of the other face image clusters; For each face image cluster in the other face image clusters, normalization processing is performed based on the number of face images corresponding to each face image cluster and the total number of the other face image clusters to obtain a normalized value of each face image cluster in the other face image clusters, wherein a larger normalized value indicates that the face image cluster is closer to the actual user face image; A first number of second human face image clusters are determined from the other human face image clusters in descending order of the normalized values.

14. The identity verification device according to claim 11, characterized in that: The determining unit determines the facial image used for the identity authentication based on the first number of second facial image clusters in the following manner: Central facial images corresponding to some second facial image clusters are selected from the first number of second facial image clusters as facial images for performing the identity authentication.

15. An identity verification device, characterized in that: The identity verification device comprises: a memory for storing instructions; and A processor, configured to call the instructions stored in the memory to execute the identity authentication method as described in any one of claims 1 to 3 or the identity authentication method as described in any one of claims 4 to 7.

16. A computer-readable storage medium storing instructions, wherein when the instructions are executed by a processor, the identity authentication method according to any one of claims 1 to 3 or the identity authentication method according to any one of claims 4 to 7 is executed.

Images