Virtual resource account processing method and device

Abstract

Embodiments of the present specification provide methods and devices for processing virtual resource accounts, wherein a method for processing a virtual resource account includes: obtaining feature similarity between a multimodal feature of the user and a stored benchmark multimodal feature according to an account application request submitted by the user for a virtual world; performing identity verification on the user based on the feature similarity, and collecting modal data of at least one modality after the verification is passed; performing secondary identity verification on the user based on data features of the modal data and benchmark data features of a corresponding modality in the benchmark multimodal feature; and creating a virtual resource account for the user in the virtual world after the secondary verification is passed.

Description

Technical Field

[0001] This document relates to the field of virtualization technology, and in particular to a method and device for processing virtual resource accounts. Background Art

[0002] With the continuous development of Internet technology, the virtual world has gradually become a hot topic in academia and industry. The virtual world is a simulation of the real world. In the virtual world, users can log in to the three-dimensional virtual world through a specific identity, and engage in various activities similar to those in the real world, or even engage in activities that cannot be performed in the real world. In the virtual world, users usually use their respective user roles to carry out activities. Summary of the invention

[0003] One or more embodiments of the present specification provide a method for processing a virtual resource account, comprising: obtaining the feature similarity between the multimodal features of the user and the stored benchmark multimodal features according to an account application request submitted by the user for the virtual world. Authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed. Performing a secondary identity authentication on the user based on the data features of the modal data and the benchmark data features of the corresponding modality in the benchmark multimodal features. Creating a virtual resource account for the user in the virtual world after the secondary authentication is passed.

[0004] One or more embodiments of the present specification provide a device for processing a virtual resource account, including: a similarity acquisition module, configured to obtain the feature similarity between the multimodal features of the user and the stored benchmark multimodal features according to an account application request submitted by the user for the virtual world. An identity authentication module, configured to authenticate the user based on the feature similarity, and collect modal data of at least one modality after the authentication is passed. A secondary identity authentication module, configured to perform secondary identity authentication on the user based on the data features of the modal data and the benchmark data features of the corresponding modality in the benchmark multimodal features. An account creation module, configured to create a virtual resource account for the user in the virtual world after the secondary authentication is passed.

[0005] One or more embodiments of the present specification provide a processing device for a virtual resource account, comprising: a processor; and a memory configured to store computer executable instructions, wherein when the computer executable instructions are executed, the processor: obtains the feature similarity between the multimodal features of the user and the stored benchmark multimodal features according to the account application request submitted by the user for the virtual world. Performs identity authentication on the user based on the feature similarity, and collects modal data of at least one modality after the authentication is passed. Performs secondary identity authentication on the user based on the data features of the modal data and the benchmark data features of the corresponding modality in the benchmark multimodal features. Creates a virtual resource account for the user in the virtual world after the secondary authentication is passed.

[0006] One or more embodiments of the present specification provide a storage medium for storing computer executable instructions, which implement the following process when executed by a processor: based on an account application request submitted by a user for a virtual world, obtain the feature similarity between the multimodal features of the user and the stored benchmark multimodal features. Perform identity authentication on the user based on the feature similarity, and collect modal data of at least one modality after the authentication is passed. Perform secondary identity authentication on the user based on the data features of the modal data and the benchmark data features of the corresponding modality in the benchmark multimodal features. Create a virtual resource account for the user in the virtual world after the secondary authentication is passed. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate one or more embodiments of the present specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor.

[0008] Figure 1 A processing flow chart of a method for processing a virtual resource account provided in one or more embodiments of this specification;

[0009] Figure 2 A processing flow chart of a method for processing a virtual resource account applied to an account application scenario provided by one or more embodiments of this specification;

[0010] Figure 3 A schematic diagram of a processing device for a virtual resource account provided in one or more embodiments of this specification;

[0011] Figure 4 A schematic diagram of the structure of a virtual resource account processing device provided in one or more embodiments of this specification. DETAILED DESCRIPTION

[0012] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below in conjunction with the drawings in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0013] This specification provides a method for processing a virtual resource account:

[0014] The method for processing virtual resource accounts provided in this embodiment performs identity authentication on the user based on the feature similarity between the user's multimodal features and the benchmark multimodal features, thereby ensuring the security of the account application process through identity authentication, and preventing other users from applying for an account in the name of the user, which would cause damage to the user's data privacy. After the authentication is passed, the user is further authenticated for a second time based on the collected modal data of at least one modality, and a virtual resource account for the user is created after the second authentication is passed. On the basis of the above-mentioned identity authentication, the depth of the control over the account application process is further improved through a second identity authentication. In this way, the user is authenticated based on two aspects of data, which improves the comprehensiveness of the data required for identity authentication, helps to improve the effectiveness and accuracy of the identity authentication results, and further improves the security of account creation and the effectiveness of the virtual resource account.

[0015] Reference Figure 1 The method for processing a virtual resource account provided in this embodiment specifically includes steps S102 to S108.

[0016] Step S102: acquiring feature similarity between the multimodal features of the user and the stored reference multimodal features according to the account application request submitted by the user to the virtual world.

[0017] The virtual world described in this embodiment refers to a virtual simulation world based on decentralized collaboration and with an open economic system. Specifically, users in the real world can access the virtual world through access devices to conduct behavioral activities in the virtual world, such as a game virtual world where users interact with each other through access devices, or a conference virtual world where users conduct online meetings through access devices; the access device of the virtual world can be a VR (Virtual Reality) device connected to the virtual world, an AR (Augmented Reality) device, etc., such as a head-mounted VR device connected to the virtual world.

[0018] In this embodiment, the modality includes a signal modality and / or an image modality. In addition, the modality may also include other types of data modalities, such as a voice modality. The multimodal data is composed of data of at least one modality, such as the multimodal data includes key physiological signals and / or key biometric images. In addition, the multimodal data may also include key voice data. The multimodal features include features extracted from the multimodal data. When the multimodal data includes key voice data, the features extracted from the multimodal data are voiceprint features, that is, the multimodal features may include voiceprint features. The reference multimodal features refer to comparison references or comparison bases for comparison with the multimodal features. The reference multimodal features may be pre-stored. The reference multimodal features include reference signal features, reference image features and / or reference voiceprint features.

[0019] In a specific implementation, a user may submit an account application request during access to a virtual world, that is, an account application request for applying for a virtual resource account in the virtual world. After receiving the account application request submitted by the user, the feature similarity between the user's multimodal features and the stored reference multimodal features may be obtained. Specifically, a first similarity between the user's signal features and the reference signal features may be obtained. Optionally, the signal features are obtained by extracting features from key physiological signals; and / or, a second similarity between the user's image features and the reference image features is obtained. Optionally, the image features are obtained by feature encoding key biometric feature images, and / or, a third similarity between the user's voiceprint features and the reference voiceprint features is obtained. Optionally, the voiceprint features are obtained by extracting features from key voice data. The key voice data is obtained after denoising the user's voice data collected by an access device that calls the virtual world or after filtering the voice data. In addition, the key voice data may also be obtained after preprocessing the voice data. Optionally, the preprocessing includes denoising or filtering. Optionally, the voice data is collected by a voice sensor integrated or configured by the access device of the virtual world.

[0020] It should be noted that the feature similarity between the user's multimodal features and the benchmark multimodal features can be calculated in advance. After receiving the account application request, the feature similarity is obtained according to the account application request. In this way, the time of identity authentication is saved by pre-calculating the feature similarity, thereby improving the authentication efficiency of identity authentication.

[0021] In practical applications, multimodal features can be signal features of key physiological signals or image features of key biometric images. In addition, multimodal features can also be features of other modalities, such as voiceprint features of speech modality, that is, voiceprint features of key speech data. In view of this, this embodiment provides the following three optional implementation methods for calculating feature similarity.

[0022] (1) Implementation method 1

[0023] In an optional implementation manner provided by this embodiment, the multimodal features include signal features of key physiological signals;

[0024] Accordingly, in order to provide support for the computational efficiency of feature similarity, a feature extractor may be introduced to extract features of key physiological signals, and specifically, the multimodal features include signal features of key physiological signals;

[0025] Accordingly, the feature similarity is calculated in the following way:

[0026] Inputting the key physiological signal into a feature extraction network to extract features, thereby obtaining signal features of the key physiological signal;

[0027] The similarity between the signal feature and the reference signal feature is calculated as the feature similarity.

[0028] In order to improve the accuracy of physiological signals required for calculating feature similarity and further improve the accuracy and effectiveness of feature similarity, optionally, the key physiological signals are obtained after denoising the physiological signals of the user collected by the access device that calls the virtual world or filtering the physiological signals. In addition, the key physiological signals can also be obtained after preprocessing the collected physiological signals, and the preprocessing includes denoising or filtering. The physiological signals include at least one of the following: electroencephalogram signals, respiratory signals, heart rate, and blood pressure.

[0029] The physiological signal refers to the electrical signal collected by the signal sensor when the user performs physiological activities, for example, the EEG signal refers to the electrical signal of the user's brain activity collected by the brain wave sensor; optionally, in the scenario where the user accesses the virtual world, the physiological signal is collected by the signal sensor integrated or configured by the access device of the virtual world, for example, the user's EEG signal is collected by the brain wave sensor configured by the access device.

[0030] The denoising process includes filtering out the noise in the physiological signal, and the denoising process can be specifically performed using a median filter method. The key physiological signal refers to a signal obtained by denoising the physiological signal, that is, a signal obtained after filtering out the noise in the physiological signal. The reference signal feature includes a comparison benchmark or comparison basis for feature comparison with the signal feature, such as a comparison benchmark or comparison basis for feature similarity or feature similarity comparison with the signal feature.

[0031] In addition, the calculation method of the feature similarity in the above implementation method 1 can be replaced by: determining or calculating the similarity between the signal feature and the reference signal feature extracted from the key physiological signal as the feature similarity; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0032] Alternatively, the signal feature extracted from the key physiological signal and the reference signal feature are input into the similarity calculation model to perform similarity calculation, and the similarity between the signal feature and the reference signal feature is obtained as the feature similarity; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0033] Alternatively, the key physiological signal is input into the feature extraction network for feature extraction at a preset time interval to obtain signal features; the similarity between the signal features and the reference signal features is calculated as the feature similarity, for example, if the preset time interval is 5 minutes, feature extraction is performed every 5 minutes; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0034] Alternatively, signal features are extracted from key physiological signals according to a preset time period, and the similarity between the signal features and the reference signal features is calculated as feature similarity; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment.

[0035] In practical applications, the training of the feature extraction network can be completed in advance, such as training the feature extraction network on a cloud server, or training the feature extraction network offline; in the process of training the feature extraction network, in order to improve the training efficiency, in an optional implementation manner provided in this embodiment, the feature extraction network is trained in the following manner:

[0036] Inputting the physiological signal sample pair into the feature extractor in the signal feature extraction model to be trained to extract features, and obtaining a signal feature pair;

[0037] Calculating a contrast loss according to the signal feature pair and the mapping feature pair, and adjusting parameters of the feature extractor according to the contrast loss;

[0038] Optionally, the mapping feature pair is obtained by inputting the signal feature pair into a feature mapper for feature mapping.

[0039] Optionally, the physiological signal sample pair may include a first sample and a second sample; the feature mapping includes performing feature conversion on the signal feature pair, or performing feature space mapping on the signal feature pair; the feature space mapping refers to mapping the signal feature pair to a target feature space. The signal feature extraction model to be trained includes a feature extraction network (feature extractor) and a feature mapping network (feature mapper).

[0040] It should be noted that the physiological signal sample pairs include sample pairs constructed by continuously collecting and storing physiological signals of users in a specific time period, such as continuously collecting and storing physiological signals of one or more users in the last 5 minutes to construct physiological signal sample pairs.

[0041] In addition to the above-mentioned implementation method of training the feature extractor to obtain the feature extraction network (i.e., the trained feature extractor) after the loss function converges, any one of the following two training methods can be used to train and obtain the feature extraction network:

[0042] Input the physiological signal sample pair into the feature extractor in the signal feature extraction model to be trained to extract features and obtain signal feature pairs; input the signal feature pair into the feature mapper in the signal feature extraction model to perform feature mapping and obtain mapping feature pairs; calculate the contrast loss according to the signal feature pair and the mapping feature pair, and adjust the parameters of the feature extractor according to the contrast loss; on this basis, the training method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0043] Alternatively, the physiological signal sample pair is input into the feature extractor in the signal feature extraction model to be trained for feature extraction to obtain a signal feature pair; the loss value is calculated based on the signal feature pair and the sample label of the physiological signal sample pair, and the parameters of the feature extractor are adjusted according to the loss value; on this basis, the training method provided here can be combined with other processing steps provided in this embodiment to form a new implementation method.

[0044] In the above process of calculating the contrast loss according to the signal feature pair and the mapping feature pair, in an optional implementation manner provided by this embodiment, the following operations are performed:

[0045] calculating a first contrast loss based on a first signal feature in the signal feature pair and a second mapping feature in the mapping feature pair, and calculating a second contrast loss based on a second signal feature in the signal feature pair and a first mapping feature in the mapping feature pair;

[0046] Calculating the sum of the first contrast loss and the second contrast loss as the contrast loss;

[0047] Optionally, the first signal feature and the first mapping feature correspond to a first sample, and the second signal feature and the second mapping feature correspond to a second sample.

[0048] Referring to the above training method, the feature extractor in the signal feature extraction model to be trained is trained by repeating the above training process, and the feature extractor in the signal feature extraction model to be trained can be adjusted with the help of a feature mapper. The feature extractor can also be adjusted using SGD (Stochastic Gradient Descent) until the loss function converges. After the loss function converges, the training is completed, and the feature extractor obtained through training is used as a feature extraction network.

[0049] For example, in the process of training the feature extractor in the signal feature extraction model to be trained, the first sample included in the physiological signal sample pair is A, and the second sample included is B, and the constructed loss function is:

[0050]

[0051] in, is the first signal feature corresponding to the first sample A, is the first mapping feature corresponding to the first sample A, is the second signal feature corresponding to the second sample B, is the second mapping feature corresponding to the second sample B, is the first contrast loss or the feature similarity or feature distance (such as Euclidean distance) between the first signal feature and the second mapping feature, is the second contrast loss or the feature similarity or feature distance (such as Euclidean distance) between the second signal feature and the first mapping feature. keguan The sum of the first contrast loss and the second contrast loss or the sum of the first feature similarity and the second feature similarity is the contrast loss. Through this loss function, the contrast loss of the signal feature pair and the mapping feature pair can be calculated.

[0052] In addition, the above-mentioned process of calculating the contrast loss based on the signal feature pair and the mapping feature pair can be replaced by calculating the first feature similarity based on the first signal feature in the signal feature pair and the second mapping feature in the mapping feature pair, and calculating the second feature similarity based on the second signal feature in the signal feature pair and the first mapping feature in the mapping feature pair; calculating the sum of the first feature similarity and the second feature similarity as the contrast loss; wherein the first signal feature and the first mapping feature correspond to the first sample, and the second signal feature and the second mapping feature correspond to the second sample; on this basis, the loss calculation process provided here can form a new implementation method with other processing steps provided in this embodiment.

[0053] (2) Implementation method 2

[0054] In an optional implementation manner provided by this embodiment, the multimodal features include image features of key biometric feature images;

[0055] Accordingly, in order to support the calculation efficiency of feature similarity, a feature encoder can be introduced to perform feature encoding of key physiological signals. Specifically, the feature similarity is calculated and obtained in the following manner:

[0056] Inputting the key biometric feature image into a feature coding network for feature coding to obtain image features of the key biometric feature image;

[0057] The similarity between the image feature and the reference image feature is calculated as the feature similarity.

[0058] Wherein, the reference image feature includes a comparison reference or comparison basis for feature comparison with the image feature, such as a comparison reference or comparison basis for feature similarity comparison or feature similarity comparison with the image feature; the feature encoder includes an image feature extractor, and accordingly, the image feature encoding includes image feature extraction. Optionally, the key biometric image includes a biometric image obtained after image evaluation of the user's biometric image or a biometric image obtained by screening in the biometric image based on image indicators; the image evaluation includes calculating image indicators, such as image indicators are image quality scores.

[0059] In addition, the calculation method of the feature similarity in the above-mentioned implementation method 2 can be replaced by: according to the reference image feature and the image feature obtained by feature encoding the key biometric feature image, determining or calculating the similarity between the image feature and the reference image feature as the feature similarity; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0060] Alternatively, the reference image features and the image features obtained by feature encoding the key biometric feature image are input into the similarity calculation model for similarity calculation, and the similarity between the reference image features and the image features is obtained as the feature similarity; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0061] Alternatively, based on the trigger event, the key biometric feature image is input into the feature encoding network for feature encoding to obtain the image feature; the similarity between the image feature and the reference image feature is calculated as the feature similarity, and optionally, the trigger event includes an identity authentication event, such as an identity authentication process during the payment operation, that is, an identity authentication event; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0062] Alternatively, the key biometric image is feature encoded based on the trigger event, and the similarity between the encoded image features and the reference image features is calculated as the feature similarity; on this basis, the calculation method provided here can form a new implementation method with other processing steps provided in this embodiment.

[0063] Similar to the above-mentioned feature extraction network, the feature encoding network can also be obtained by pre-training, such as training the feature encoding network on a cloud server, or training the feature encoding network offline; in the process of training the feature encoding network, in order to improve the training efficiency, in an optional implementation manner provided in this embodiment, the feature encoding network is trained in the following manner:

[0064] Inputting the biometric image sample into the feature encoder of the image feature encoding model to be trained for feature encoding to obtain sample image features;

[0065] Calculating a loss value based on the identity category probability and the sample label corresponding to the biometric image sample;

[0066] Adjusting parameters of the feature encoder according to the loss value;

[0067] Optionally, the identity category probability is obtained by inputting the sample image features into a probability calculator for probability calculation; the identity category probability includes the probability of characterizing the user identity, and the identity category probability includes at least one probability number, such as the sample label of the biometric image sample is marked with user a and user b, and the identity category probability includes the probability that the sample image is user a and the probability that the sample image is user b. The image feature encoding model to be trained includes a feature encoding network (feature encoder) and a probability calculation network (probability calculator).

[0068] It should be noted that the biometric image samples include biometric images of users collected according to a preset collection period and storing biometric images of a specific time period. For example, biometric images of one or more users are collected every 30 seconds and biometric images of the last 5 minutes are stored to construct biometric image samples.

[0069] Referring to the above training method, the feature encoder is trained by repeating the above training process, and the feature encoder can be adjusted with the help of a probability calculator, or SGD (Stochastic Gradient Descent) can be used to adjust the parameters of the feature encoder until the loss function converges. After the loss function converges, the training is completed, and the feature encoder obtained by training is used as the feature encoding network.

[0070] For example, in the process of training the feature encoder in the image feature encoding model to be trained, the loss function constructed is:

[0071] Loss jishi =Softmax(f,y)

[0072] Among them, f is the identity category probability, y is the sample probability, Softmax(f, y) is the loss value calculated based on the identity category probability f and the sample probability y. The loss value is used to adjust the parameters until the loss function converges to obtain the feature encoding network.

[0073] In addition to the above-mentioned implementation method of training the feature encoder to obtain the feature extraction network after the loss function converges, any one of the following three training methods can be used to train and obtain the feature extraction network: input the biometric image sample into the feature encoder in the image feature encoding model to be trained for feature encoding to obtain sample image features; input the sample image features into the probability calculator in the image feature encoding model for probability calculation to obtain identity category probability; calculate the loss value according to the identity category probability and the sample probability corresponding to the biometric image sample, and adjust the parameters of the feature encoder according to the loss value; on this basis, the training method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0074] Alternatively, the biometric image sample is input into a feature encoder in an image feature coding model to be trained for feature coding to obtain sample image features; a loss value is calculated according to the identity category probability and the sample probability corresponding to the biometric image sample; parameters of the image feature coding model to be trained are adjusted according to the loss value; optionally, the identity category probability is obtained by inputting the sample image features into a probability calculator for probability calculation; on this basis, the training method provided herein can form a new implementation method with other processing steps provided in this embodiment;

[0075] Alternatively, the biometric image sample is input into a feature encoder in an image feature encoding model to be trained for feature encoding to obtain sample image features; a loss value is calculated based on the identity classification result and the sample label corresponding to the biometric image sample; and parameters of the feature encoder are adjusted based on the loss value; optionally, the identity classification result is obtained by inputting the sample image feature into a classifier for identity category classification, the identity classification result includes identity category probability, and the sample label includes sample probability; on this basis, the training method provided herein can be combined with other processing steps provided in this embodiment to form a new implementation method.

[0076] In order to improve the accuracy of the biometric image required for calculating the feature similarity and further improve the accuracy and effectiveness of the feature similarity, a biometric image can be selected from the user's biometric image based on the image index as the key biometric image. Specifically, in an optional implementation provided by this embodiment, the key biometric image is obtained by:

[0077] Inputting the user's biometric image into an index calculation model to perform image index calculation to obtain an image index of the biometric image;

[0078] The biometric image whose image index is greater than the index threshold is used as the key biometric image.

[0079] The biometric image refers to an image containing the user's biometrics collected by an image sensor, and the biometric image includes a facial image and / or an eye image. Optionally, in a scenario where a user accesses a virtual world, the biometric image is collected by an image sensor integrated or configured in an access device of the virtual world; the image index includes an image quality score.

[0080] In addition, the above-mentioned method of obtaining the key biometric image can be replaced by performing an image evaluation on the user's biometric image, and screening a biometric image from the biometric images as the key biometric image based on the image evaluation result; on this basis, the method of obtaining the key biometric image provided here can form a new implementation method with other processing steps provided in this embodiment; optionally, the image evaluation includes calculating image indicators, and the image indicators include image quality scores.

[0081] Alternatively, an image index of the biometric image is calculated according to the biometric image of the user, and a key biometric image is determined in the biometric image based on the calculated image index; optionally, the determining of the key biometric image in the biometric image based on the calculated image index includes taking a biometric image whose image index is greater than an index threshold as the key biometric image; on this basis, the method for obtaining the key biometric image provided herein can form a new implementation method with other processing steps provided in this embodiment;

[0082] Alternatively, the image index of the biometric image is calculated based on the biometric image of the user, and the biometric image is selected from the biometric image based on the image index obtained by calculation as the key biometric image; on this basis, the method for obtaining the key biometric image provided herein can form a new implementation method with other processing steps provided in this embodiment;

[0083] Alternatively, the image index of the biometric image is calculated based on the user's biometric image, and the biometric image whose image index is greater than the index threshold is used as the key biometric image; on this basis, the method for obtaining the key biometric image provided here can form a new implementation method with other processing steps provided in this embodiment.

[0084] (3) Implementation method 3

[0085] In order to improve the comprehensiveness of multimodal features and further improve the comprehensiveness of feature similarity, the signal similarity between the signal feature and the reference signal feature and the image similarity between the image feature and the reference image feature may be simultaneously obtained as the feature similarity.

[0086] Specifically, the multimodal features may also include signal features of key physiological signals and image features of key biometric images; the feature similarity between the multimodal features and the stored benchmark multimodal features may be calculated in the following manner: inputting the key physiological signals into a feature extraction network for feature extraction to obtain signal features, and inputting the key biometric images into a feature encoding network for feature encoding to obtain image features; calculating the signal similarity between the signal features and the benchmark signal features, and calculating the image similarity between the image features and the benchmark image features; and using the signal similarity and the image similarity as the feature similarity.

[0087] It should be supplemented that the above step S102 can be replaced by obtaining a feature index corresponding to the multimodal feature of the user according to an account application request submitted by the user for the virtual world; optionally, the feature index includes a feature similarity between the multimodal feature and a stored benchmark multimodal feature, and constitutes a new implementation method together with other processing steps provided in this embodiment.

[0088] Step S104: authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed.

[0089] The above-mentioned process obtains the feature similarity between the user's multimodal features and the stored benchmark multimodal features based on the account application request submitted by the user for the virtual world. In this step, the user is authenticated based on the feature similarity, and corresponding processing is performed based on the identity verification result. Specifically, if the authentication is successful, modal data of at least one modality is collected; if the authentication fails, an authentication failure reminder is output to the user's access device; in addition, if the authentication is successful, the stored benchmark multimodal features can also be updated based on the above-mentioned multimodal features to improve the accuracy of the comparison benchmark.

[0090] The identity authentication described in this embodiment refers to verifying or checking the identity of the user; optionally, the at least one modality is composed of an image modality, a signal modality and / or a voice modality, and accordingly, the modality data includes current physiological signals, current biometric images and / or current voice data, such as collecting the user's current EEG signals or current facial images.

[0091] In specific implementation, the user is authenticated with the help of feature similarity to confirm the user's identity, so as to ensure the security of the account application process and prevent the user from using the account for non-compliant purposes after applying for it. After the authentication is passed, modal data of at least one modality is collected for secondary identity authentication to further enhance the security of the account application. Specifically, after the authentication is passed, the user's current biometric image, current physiological signal and / or current voice data can be collected.

[0092] In practical applications, since there may be more than one multimodal feature, there may also be more than one feature similarity. In order to improve the effectiveness of identity authentication and the convenience of identity authentication, the authentication result of identity authentication can be determined based on the number of similarities whose feature similarities are greater than the similarity threshold. Specifically, if the number of similarities is within the number value range, it is determined that the authentication is passed; if the similarity is not within the number value range, it is determined that the authentication is not passed. In an optional implementation provided by this embodiment, the user is authenticated in the following manner:

[0093] Determining the number of similarities whose feature similarity is greater than a similarity threshold;

[0094] Determine whether the similarity number is within a number value range;

[0095] If yes, it is determined that the authentication is passed; if no, it is determined that the authentication is not passed or an authentication failure reminder is output to the user's access device.

[0096] In addition, the above-mentioned implementation process of user identity authentication can be replaced by: determining the number of similarities whose feature similarity is greater than the similarity threshold; if the similarity number is within the number value range, determining that the authentication is successful; if the similarity is not within the number value range, determining that the authentication is failed or outputting an authentication failure reminder to the user's access device; on this basis, the identity authentication method provided here and the other processing steps provided in this embodiment constitute a new implementation method.

[0097] In addition, based on the third implementation method of calculating feature similarity provided above, in order to improve the accuracy of identity authentication, the authentication result can be determined based on the signal similarity and the image similarity respectively. Specifically, the user can be authenticated in the following manner: determine a first similarity number in which the signal similarity is greater than the signal similarity threshold, and determine a second similarity number in which the image similarity is greater than the image similarity threshold; if the first similarity number is within a first number value range and the second similarity number is within a second number value range, determine that the authentication is passed; wherein the first number value range and the second number value range may be the same or different.

[0098] Step S106: performing secondary identity authentication on the user based on the data features of the modal data and the benchmark data features of the corresponding modality in the benchmark multimodal features.

[0099] The above-mentioned user identity authentication is performed with the help of the feature similarity between the multimodal features and the stored benchmark multimodal features, and the modal data of at least one modality is collected after the authentication is passed. In this step, the user is authenticated for the second time with the data features of the modal data and the benchmark data features. The account application is securely checked through the second identity authentication, which can further improve the security of the account application and the effectiveness of the identity authentication, and avoid users applying for virtual resource accounts in the name of other users, thereby damaging the privacy security of other users.

[0100] The data features described in this embodiment refer to features extracted from modal data, such as data features obtained by inputting modal data into a feature extraction model for feature extraction; the corresponding modality refers to the modality corresponding to the above-mentioned collected modal data, such as collecting the current biometric image of the user, then the corresponding modality is the image modality. The reference data feature refers to a comparison basis or comparison benchmark for comparison with the data feature. Specifically, when the corresponding modality is the image modality, the reference data feature is the reference image feature, and when the corresponding modality is the signal modality, the reference data feature is the reference signal feature.

[0101] In specific implementation, in order to ensure the security of the account application process, the user can be subjected to secondary identity authentication. Specifically, the similarity between the current image features and the reference image features can be calculated based on the current image features extracted from the user's current biometric image. When the similarity meets the threshold condition, it is determined that the secondary authentication is passed. In an optional implementation provided by this embodiment, in the process of secondary identity authentication of the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features, the following operations are performed:

[0102] Extracting current image features from the current biometric image of the user, and calculating the similarity between the current image features and the reference image features;

[0103] If the similarity is greater than the similarity threshold, it is determined that the secondary authentication is passed;

[0104] If the similarity is less than or equal to the similarity threshold, it is determined that the secondary authentication has failed.

[0105] In addition, the implementation process of the secondary identity authentication of the user provided above can be replaced by calculating the similarity between the current signal feature and the reference signal feature based on the current signal feature extracted from the current physiological signal of the user; if the similarity is greater than the preset similarity threshold, it is determined that the secondary authentication is passed; on this basis, the secondary identity authentication method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0106] Alternatively, based on the current voiceprint feature extracted from the user's current voice data, the similarity between the current voiceprint feature and the reference voiceprint feature is calculated; if the similarity is greater than a preset similarity threshold, it is determined that the secondary authentication is passed; on this basis, the secondary identity authentication method provided here can form a new implementation method with other processing steps provided in this embodiment;

[0107] Alternatively, in order to improve the comprehensiveness and accuracy of the secondary identity authentication, in the process of performing secondary identity authentication on the user based on the data features of the modal data and the benchmark data features of the corresponding modality in the benchmark multimodal features, data features can also be extracted from the modal data (specifically, any two or three of the current image features of the current biometric image, the current signal features of the current physiological signal, and the current voiceprint features of the current voice information), and the similarity between the data features and the benchmark data features (corresponding to the aforementioned data features, specifically, any two or three of the benchmark image features, the benchmark signal features, and the benchmark voiceprint features) is calculated; if the similarity is greater than the similarity threshold, it is determined that the secondary authentication is passed; if the similarity is less than or equal to the similarity threshold, it is determined that the secondary authentication is not passed. Specifically, if it is any two or three, the similarity is two or three, and the similarity is greater than the similarity threshold (the similarity thresholds corresponding to the two or three similarities can be the same or different) to determine that the secondary authentication is passed; on this basis, the secondary identity authentication method provided herein can form a new implementation method with other processing steps provided in this embodiment.

[0108] Step S108: creating a virtual resource account for the user in the virtual world after the secondary authentication is passed.

[0109] In the above, the user is authenticated for a second time based on the data features of the modal data and the benchmark data features. In this step, corresponding processing is performed based on the result of the second authentication. Specifically, if the second authentication passes, a virtual resource account of the user in the virtual world is created. If the second authentication fails, an authentication failure reminder or a re-authentication message is output to the user's access device. In addition, if the second authentication passes, the benchmark data features of the corresponding modality in the benchmark multimodal features can also be updated based on the data features of the modal data.

[0110] In specific implementation, in order to improve the convenience of creating accounts for users in the real world, in an optional implementation manner provided by this embodiment, after creating a virtual resource account for the user in the virtual world after the secondary authentication is passed, the following operations are also performed: resource creation processing is performed based on the virtual resource account; accordingly, the resource creation processing based on the virtual resource account includes: creating a resource account in the real world that is mapped to the virtual resource account.

[0111] The virtual resource account includes a virtual resource account used for resource transactions in the virtual world, such as a virtual financial account or a virtual accounting account. In actual applications, no actual funds may be deposited in the virtual resource account, and the account is only used to reflect the amount of funds in the user's resource account in the real world. The real world refers to the physical world corresponding to the virtual world, that is, the real physical world; the resource account includes a real financial account or a physical financial account that has a mapping relationship with the virtual financial account in the real world.

[0112] Optionally, the user conducts resource transactions in the virtual world through the virtual resource account, and after the transaction is completed, the resource amount of the resource account in the real world is synchronized based on the resource amount of the virtual resource account; optionally, the resource amount synchronization is performed based on a mapping relationship between the resource account and the virtual resource account.

[0113] It should be noted that step S108 can be replaced by creating a virtual resource account for the user in the virtual world after the secondary authentication is passed, so as to perform resource creation processing based on the virtual resource account; optionally, the resource creation processing based on the virtual resource account includes: creating a resource account in the real world mapped to the virtual resource account, and forming a new implementation method with other processing steps provided in this embodiment.

[0114] The method for processing virtual resource accounts provided in this embodiment can be applied to a server, where the server refers to a server or service platform that provides corresponding services for access devices accessing a virtual world, or a server or service platform that maintains the operation of a virtual world.

[0115] In summary, the method for processing a virtual resource account provided in this embodiment first obtains a first similarity between a signal feature of the user and a stored reference signal feature after the user submits an account application request for the virtual world, and obtains a second similarity between an image feature of the user and a stored reference image feature, and uses the first similarity and the second similarity as feature similarity;

[0116] Secondly, determine the number of similarities whose feature similarity is greater than the similarity threshold. If the similarity number is within the number value range, determine that the identity authentication is passed, so as to ensure the security of the account application process through identity authentication, and prevent other users from applying for an account in the name of the user, resulting in damage to the user's data privacy; further collect the user's current biometric image, and perform a secondary identity authentication on the user based on the current image features and the reference image features extracted from the current biometric image; on the basis of the above identity authentication, further improve the depth of the control of the account application process through secondary identity authentication;

[0117] Finally, after the second authentication is passed, a virtual resource account for the user in the virtual world is created. In this way, the user is authenticated based on two aspects of data, which improves the comprehensiveness of the data required for authentication, helps to improve the effectiveness and accuracy of the authentication results, and further improves the security of account creation and the effectiveness of virtual resource accounts.

[0118] The following uses the application of a virtual resource account processing method provided by this embodiment in an account application scenario as an example to further illustrate the virtual resource account processing method provided by this embodiment. Figure 2 , a method for processing a virtual resource account applied to an account application scenario specifically includes the following steps.

[0119] Step S202, according to the account application request submitted by the user to the virtual world, obtain a first similarity between the signal feature of the key physiological signal of the user and the stored reference signal feature, and a second similarity between the image feature of the key biometric image and the stored reference image feature.

[0120] Step S204: taking the first similarity and the second similarity as feature similarities, and determining the number of similarities whose feature similarities are greater than a similarity threshold.

[0121] Step S206: If the similarity number is within the number value range, it is determined that the authentication result of the user identity authentication is authentication passed.

[0122] Step S208: calling the access device of the virtual world to collect the current biometric image of the user.

[0123] Step S210: extracting current image features from the current biometric image, and calculating the similarity between the current image features and the reference image features.

[0124] Step S212: If the similarity is greater than the current similarity threshold, it is determined that the result of the secondary identity authentication for the user is that the secondary authentication is passed.

[0125] Step S214: Create a virtual resource account of the user in the virtual world to create a resource account in the real world that is mapped to the virtual resource account.

[0126] The above steps S202 to S204 can be replaced by: according to the account application request submitted by the user for the virtual world, obtaining the feature similarity between the voiceprint feature of the user's key voice data and the stored reference voiceprint feature; determining the number of similarities where the feature similarity is greater than the similarity threshold, and forming a new implementation method with other processing steps provided in this embodiment;

[0127] Alternatively, based on an account application request submitted by a user for a virtual world, a first similarity between a signal feature of a key physiological signal of the user and a stored benchmark signal feature, a second similarity between an image feature of a key biometric image and a stored benchmark image feature, and a third similarity between a voiceprint feature of key voice data and a stored benchmark voiceprint feature are obtained; the first similarity, the second similarity, and the third similarity are used as feature similarities, and the number of similarities for which the feature similarity is greater than a similarity threshold is determined, and together with other processing steps provided in this embodiment, a new implementation method is formed.

[0128] The above steps S208 to S210 can be replaced by: calling the access device of the virtual world to collect the current physiological signal of the user; extracting the current signal feature from the current physiological signal, and calculating the similarity between the current signal feature and the reference signal feature, and forming a new implementation method with other processing steps provided in this embodiment;

[0129] Alternatively, the access device of the virtual world is called to collect the current physiological signal and the current biometric image of the user; the current signal feature is extracted from the current physiological signal, and the signal feature similarity between the current signal feature and the reference signal feature is calculated; and the current image feature is extracted from the current biometric image, and the image feature similarity between the current image feature and the reference image feature is calculated, and the signal feature similarity and the image feature similarity are used as the similarity in step S212, and together with other processing steps provided in this embodiment, a new implementation method is formed;

[0130] Alternatively, call the access device of the virtual world to collect the user's current voice data; extract the current voiceprint features from the current voice data, and calculate the similarity between the current voiceprint features and the baseline voiceprint features, and form a new implementation method together with other processing steps provided in this embodiment.

[0131] An embodiment of a processing device for a virtual resource account provided in this specification is as follows:

[0132] In the above-mentioned embodiment, a method for processing a virtual resource account is provided, and correspondingly, a device for processing a virtual resource account is also provided, which will be described below with reference to the accompanying drawings.

[0133] Reference Figure 3 , which shows a schematic diagram of a processing device for a virtual resource account provided in this embodiment.

[0134] Since the device embodiment corresponds to the method embodiment, the description is relatively simple, and the relevant parts can refer to the corresponding description of the method embodiment provided above. The device embodiment described below is only illustrative.

[0135] This embodiment provides a virtual resource account processing device, including:

[0136] A similarity acquisition module 302 is configured to acquire a feature similarity between a multimodal feature of a user and a stored reference multimodal feature according to an account application request submitted by the user to the virtual world;

[0137] The identity authentication module 304 is configured to authenticate the user based on the feature similarity and collect modal data of at least one modality after the authentication is passed;

[0138] A secondary identity authentication module 306 is configured to perform secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features;

[0139] The account creation module 308 is configured to create a virtual resource account of the user in the virtual world after the secondary authentication is passed.

[0140] An embodiment of a processing device for a virtual resource account provided in this specification is as follows:

[0141] Corresponding to the above-described method for processing a virtual resource account, based on the same technical concept, one or more embodiments of this specification further provide a virtual resource account processing device, the virtual resource account processing device is used to execute the above-provided method for processing a virtual resource account. Figure 4 A schematic diagram of the structure of a virtual resource account processing device provided in one or more embodiments of this specification.

[0142] This embodiment provides a virtual resource account processing device, including:

[0143] like Figure 4 As shown, the processing device of the virtual resource account may have relatively large differences due to different configurations or performances, and may include one or more processors 401 and memory 402, and one or more storage applications or data may be stored in the memory 402. Among them, the memory 402 can be a short-term storage or a persistent storage. The application stored in the memory 402 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the processing device of the virtual resource account. Further, the processor 401 can be configured to communicate with the memory 402, and execute a series of computer executable instructions in the memory 402 on the processing device of the virtual resource account. The processing device of the virtual resource account may also include one or more power supplies 403, one or more wired or wireless network interfaces 404, one or more input / output interfaces 405, one or more keyboards 406, etc.

[0144] In a specific embodiment, the processing device of the virtual resource account includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions in the processing device of the virtual resource account, and the one or more programs are configured to be executed by one or more processors, including computer executable instructions for performing the following:

[0145] According to an account application request submitted by a user to the virtual world, obtaining a feature similarity between a multimodal feature of the user and a stored reference multimodal feature;

[0146] Authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed;

[0147] Performing secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features;

[0148] After the secondary authentication is passed, a virtual resource account of the user in the virtual world is created.

[0149] An embodiment of a storage medium provided in this specification is as follows:

[0150] Corresponding to the method for processing a virtual resource account described above, based on the same technical concept, one or more embodiments of this specification further provide a storage medium.

[0151] The storage medium provided in this embodiment is used to store computer executable instructions, and the computer executable instructions implement the following process when executed by a processor:

[0152] According to an account application request submitted by a user to the virtual world, obtaining a feature similarity between a multimodal feature of the user and a stored reference multimodal feature;

[0153] Authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed;

[0154] Performing secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features;

[0155] After the secondary authentication is passed, a virtual resource account of the user in the virtual world is created.

[0156] It should be noted that the embodiment of a storage medium in this specification and the embodiment of a method for processing a virtual resource account in this specification are based on the same inventive concept, so the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0157] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0158] In the 1930s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0159] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0160] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0161] For the convenience of description, the above devices are described in terms of functions and are divided into various units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0162] It should be understood by those skilled in the art that one or more embodiments of this specification may be provided as a method, system or computer program product. Therefore, one or more embodiments of this specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0164] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0166] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0167] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0168] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0169] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0170] One or more embodiments of the present specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0171] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0172] The above description is only an embodiment of this document and is not intended to limit this document. For those skilled in the art, this document may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this document should be included in the scope of the claims of this document.

Claims

1. A method for processing a virtual resource account, comprising: According to an account application request submitted by a user for a virtual world, obtaining a feature similarity between a multimodal feature of the user and a stored benchmark multimodal feature, wherein the virtual world refers to a virtual simulated world based on decentralized collaboration and having an open economic system; Authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed; Performing secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features; After the secondary authentication is passed, a virtual resource account of the user in the virtual world is created; The multimodal features include signal features of key physiological signals, wherein the key physiological signals are obtained after preprocessing the physiological signals of the user collected by the access device that calls the virtual world, and the physiological signals refer to electrical signals collected by a signal sensor when the user performs physiological activities; Accordingly, the feature similarity is calculated in the following way: Inputting the key physiological signal into a feature extraction network to extract features, thereby obtaining signal features of the key physiological signal; Calculating the similarity between the signal feature and the reference signal feature as the feature similarity; The feature extraction network is trained in the following way: Inputting the physiological signal sample pair into the feature extractor in the signal feature extraction model to be trained to extract features, and obtaining a signal feature pair; Calculating a contrast loss according to the signal feature pair and the mapping feature pair, and adjusting parameters of the feature extractor according to the contrast loss; Wherein, the mapping feature pair is obtained by inputting the signal feature pair into a feature mapper for feature mapping; The authenticating the user based on the feature similarity includes: Determining the number of similarities whose feature similarity is greater than a similarity threshold; Determine whether the similarity number is within a number value range; If yes, confirm that the authentication is successful.

2. According to the method for processing a virtual resource account according to claim 1, the physiological signal comprises at least one of the following: an electroencephalogram signal, a respiratory signal, a heart rate, and blood pressure.

3. The method for processing a virtual resource account according to claim 1, wherein the step of calculating the contrast loss based on the signal feature pair and the mapping feature pair comprises: calculating a first contrast loss based on a first signal feature in the signal feature pair and a second mapping feature in the mapping feature pair, and calculating a second contrast loss based on a second signal feature in the signal feature pair and a first mapping feature in the mapping feature pair; Calculating the sum of the first contrast loss and the second contrast loss as the contrast loss; The first signal feature and the first mapping feature correspond to a first sample, and the second signal feature and the second mapping feature correspond to a second sample.

4. The method for processing a virtual resource account according to claim 1, wherein the multimodal feature comprises an image feature of a key biometric feature image; Accordingly, the feature similarity is calculated in the following way: Inputting the key biometric feature image into a feature coding network for feature coding to obtain image features of the key biometric feature image; The similarity between the image feature and the reference image feature is calculated as the feature similarity.

5. According to the method for processing a virtual resource account of claim 4, the key biometric feature image is obtained by: Inputting the user's biometric image into an index calculation model to perform image index calculation to obtain an image index of the biometric image; The biometric image whose image index is greater than the index threshold is used as the key biometric image.

6. According to the method for processing virtual resource accounts of claim 4, the feature encoding network is trained in the following manner: Inputting the biometric image sample into the feature encoder of the image feature encoding model to be trained for feature encoding to obtain sample image features; Calculating a loss value based on the identity category probability and the sample label corresponding to the biometric image sample; Adjusting parameters of the feature encoder according to the loss value; in, The identity category probability is obtained by inputting the sample image features into a probability calculator to perform probability calculation.

7. The method for processing a virtual resource account according to claim 1, wherein the second authentication of the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features comprises: Extracting current image features from the current biometric image of the user, and calculating the similarity between the current image features and the reference image features; If the similarity is greater than the similarity threshold, it is determined that the secondary authentication is passed.

8. The method for processing a virtual resource account according to claim 1, further comprising: Perform resource creation processing based on the virtual resource account; Accordingly, the resource creation process based on the virtual resource account includes: A resource account is created in the real world that is mapped to the virtual resource account.

9. A virtual resource account processing device, comprising: a similarity acquisition module configured to acquire, according to an account application request submitted by a user for a virtual world, a feature similarity between a multimodal feature of the user and a stored reference multimodal feature, wherein the virtual world refers to a virtual simulated world based on decentralized collaboration and having an open economic system; An identity authentication module, configured to authenticate the user based on the feature similarity, and collect modal data of at least one modality after the authentication is passed; A secondary identity authentication module, configured to perform secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features; An account creation module, configured to create a virtual resource account of the user in the virtual world after the secondary authentication is passed; The multimodal features include signal features of key physiological signals, wherein the key physiological signals are obtained after denoising the physiological signals of the user collected by the access device that calls the virtual world, and the physiological signals refer to electrical signals collected by a signal sensor when the user performs physiological activities; Accordingly, the feature similarity is calculated in the following way: Inputting the key physiological signal into a feature extraction network to extract features, thereby obtaining signal features of the key physiological signal; Calculating the similarity between the signal feature and the reference signal feature as the feature similarity; The feature extraction network is trained in the following way: Inputting the physiological signal sample pair into the feature extractor in the signal feature extraction model to be trained to extract features, and obtaining a signal feature pair; Calculating a contrast loss according to the signal feature pair and the mapping feature pair, and adjusting parameters of the feature extractor according to the contrast loss; Wherein, the mapping feature pair is obtained by inputting the signal feature pair into a feature mapper for feature mapping; The authenticating the user based on the feature similarity includes: Determining the number of similarities whose feature similarity is greater than a similarity threshold; Determine whether the similarity number is within a number value range; If yes, confirm that the authentication is successful.

10. A virtual resource account processing device, comprising: processor; and a memory configured to store computer executable instructions that, when executed, cause the processor to: According to an account application request submitted by a user for a virtual world, obtaining a feature similarity between a multimodal feature of the user and a stored benchmark multimodal feature, wherein the virtual world refers to a virtual simulated world based on decentralized collaboration and having an open economic system; Authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed; Performing secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features; After the secondary authentication is passed, a virtual resource account of the user in the virtual world is created; The multimodal features include signal features of key physiological signals, wherein the key physiological signals are obtained after denoising the physiological signals of the user collected by the access device that calls the virtual world, and the physiological signals refer to electrical signals collected by a signal sensor when the user performs physiological activities; Accordingly, the feature similarity is calculated in the following way: Inputting the key physiological signal into a feature extraction network to extract features, thereby obtaining signal features of the key physiological signal; Calculating the similarity between the signal feature and the reference signal feature as the feature similarity; The feature extraction network is trained in the following way: Inputting the physiological signal sample pair into the feature extractor in the signal feature extraction model to be trained to extract features, and obtaining a signal feature pair; Calculating a contrast loss according to the signal feature pair and the mapping feature pair, and adjusting parameters of the feature extractor according to the contrast loss; Wherein, the mapping feature pair is obtained by inputting the signal feature pair into a feature mapper for feature mapping; The authenticating the user based on the feature similarity includes: Determining the number of similarities whose feature similarity is greater than a similarity threshold; Determine whether the similarity number is within a number value range; If yes, confirm that the authentication is successful.

11. A storage medium for storing computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the following process: According to an account application request submitted by a user for a virtual world, obtaining a feature similarity between a multimodal feature of the user and a stored benchmark multimodal feature, wherein the virtual world refers to a virtual simulated world based on decentralized collaboration and having an open economic system; Authenticating the user based on the feature similarity, and collecting modal data of at least one modality after the authentication is passed; Performing secondary identity authentication on the user based on the data features of the modal data and the reference data features of the corresponding modality in the reference multimodal features; After the secondary authentication is passed, a virtual resource account of the user in the virtual world is created; The multimodal features include signal features of key physiological signals, wherein the key physiological signals are obtained after denoising the physiological signals of the user collected by the access device that calls the virtual world, and the physiological signals refer to electrical signals collected by a signal sensor when the user performs physiological activities; Accordingly, the feature similarity is calculated in the following way: Inputting the key physiological signal into a feature extraction network to extract features, thereby obtaining signal features of the key physiological signal; Calculating the similarity between the signal feature and the reference signal feature as the feature similarity; The feature extraction network is trained in the following way: Inputting the physiological signal sample pair into the feature extractor in the signal feature extraction model to be trained to extract features and obtain signal feature pairs; Calculating a contrast loss according to the signal feature pair and the mapping feature pair, and adjusting parameters of the feature extractor according to the contrast loss; The mapping feature pair is obtained by inputting the signal feature pair into a feature mapper for feature mapping; The authenticating the user based on the feature similarity includes: Determining the number of similarities whose feature similarity is greater than a similarity threshold; Determine whether the similarity number is within a number value range; If yes, confirm that the authentication is successful.

Images