Systems, methods, and computer program products for evolutionary learning in verification template matching during biometric authentication

Abstract

Systems for authenticating an individual using image feature templates are provided, the systems including at least one processor to train a first machine learning model based on a training data set of a plurality of images of a user, generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the user during a time interval, generate a second machine learning model based on the plurality of image feature templates, generate a predicted image feature template using the second machine learning model, determine whether to authenticate the identity of the user based on an input image of the user, and perform an action based on determining whether to authenticate the identity of the user. Methods and computer program products are also provided.

Description

[0001] Cross Reference to Related Applications

[0002] This application claims priority to U.S. Patent Application No. 17 / 118,740, filed December 11, 2020, the disclosure of which is hereby incorporated by reference in its entirety. TECHNICAL FIELD

[0003] The present disclosure relates generally to biometric authentication, and in some non-limiting embodiments or aspects, to systems, methods, and computer program products for authenticating individuals using templates developed based on evolutionary learning. BACKGROUND

[0004] Authentication refers to a process performed electronically that involves proving an assertion, such as verifying (e.g., proving) that an individual’s identity is correct. Authentication can be contrasted with identification, which refers to a process of determining and / or indicating a user’s identity. Biometric authentication can refer to the use of a unique physical characteristic (e.g., a body measurement) of an individual as an intermediary for authentication. In some cases, physical characteristics that can be used for authentication include fingerprints, voice samples, facial images, and / or eye images (e.g., iris scans) because these physical characteristics are unique to an individual. In some cases, machine learning can be used in conjunction with biometric authentication because machine learning enables processes associated with authentication to be driven based on the need to maintain large sets of analyzed parameters and data based on biometric authentication.

[0005] However, machine learning models can not account for changes in physical characteristics of an individual over time. For example, a machine learning model can not be able to authenticate an individual based on facial images of the individual when aspects of an initial facial image of the individual differ from a later facial image of the individual due to aging. In some cases, many samples of physical characteristics of an individual can be taken over time. However, obtaining and storing many samples can require a large amount of technical resources. SUMMARY

[0006] Accordingly, improved systems, methods, and computer program products are disclosed for authenticating individuals using feature templates developed based on evolutionary learning.

[0007] According to some non-limiting embodiments or aspects, a system is provided, comprising: at least one processor programmed or configured to: train a first machine learning model based on a training dataset of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users within a time interval; generate a second machine learning model based on the plurality of image feature templates; generate a predicted image feature template using the second machine learning model; determine whether to authenticate an identity of a second user based on an input image of the second user, wherein when determining whether to authenticate the identity of the second user based on the input image of the second user, the at least one processor is programmed or configured to: generate a current image feature template based on the input image of the second user, compare the current image feature template to the predicted image feature template, and determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on determining whether to authenticate the identity of the second user.

[0008] According to some non-limiting embodiments or aspects, a method is provided, comprising: training, with at least one processor, a first machine learning model based on a training dataset of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generating, with at least one processor, a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users within a time interval; generating, with at least one processor, a second machine learning model based on the plurality of image feature templates; generating, with at least one processor, a predicted image feature template using the second machine learning model; determining, with at least one processor, whether to authenticate an identity of a second user based on an input image of the second user, wherein determining whether to authenticate the identity of the second user based on the input image of the second user comprises: generating a current image feature template based on the input image of the second user, comparing the current image feature template to the predicted image feature template, and determining whether the current image feature template corresponds to the predicted image feature template; and performing an action based on determining whether to authenticate the identity of the second user.

[0009] According to some non-limiting embodiments or aspects, a computer program product is provided that includes at least one non-transitory computer-readable medium comprising one or more instructions that, when executed by at least one processor, cause the at least one processor to: train a first machine learning model based on a training data set of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users over a time interval; generate a second machine learning model based on the plurality of image feature templates; generate a predicted image feature template using the second machine learning model; determine to authenticate an identity of a second user based on an input image of the second user, wherein the one or more instructions that cause the at least one processor to determine to authenticate the identity of the second user based on the input image of the second user cause the at least one processor to: generate a current image feature template based on the input image of the second user, compare the current image feature template to the predicted image feature template, and determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on determining whether to authenticate the identity of the second user.

[0010] Other non-limiting embodiments or aspects are set forth in the following numbered clauses:

[0011] Clause 1 : A system comprising: at least one processor programmed or configured to: train a first machine learning model based on a training data set of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users over a time interval; generate a second machine learning model based on the plurality of image feature templates; generate a predicted image feature template using the second machine learning model; determine to authenticate an identity of a second user based on an input image of the second user, wherein, when determining to authenticate the identity of the second user based on the input image of the second user, the at least one processor is programmed or configured to: generate a current image feature template based on the input image of the second user, compare the current image feature template to the predicted image feature template, and determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on determining whether to authenticate the identity of the second user.

[0012] Clause 2: The system of clause 1, wherein, when training the first machine learning model based on the training data set of a plurality of images of the one or more first users, the at least one processor is programmed or configured to: train the first machine learning model based on a training data set of a plurality of facial images of the one or more first users.

[0013] Clause 3: The system of clause 1 or 2, wherein, when generating the plurality of image feature templates using the first machine learning model, the at least one processor is programmed or configured to: extract a first image feature template from the first machine learning model after training the first machine learning model.

[0014] Clause 4: The system of any of clauses 1-3, wherein the at least one processor is programmed or configured to: add the input image of the second user to the plurality of images of the one or more first users in the training data set to provide an updated training data set; and retrain the first machine learning model based on the updated training data set.

[0015] Clause 5: The system of any of clauses 1-4, wherein, when generating the plurality of image feature templates using the first machine learning model, the at least one processor is programmed or configured to: extract a first image feature template from the first machine learning model after retraining the first machine learning model based on the updated training data set.

[0016] Clause 6: The system of any of clauses 1-5, wherein the first machine learning model is a convolutional neural network model and the second machine learning model is a long short-term memory recurrent neural network model.

[0017] Clause 7: The system of clauses 1-6, wherein the predicted image feature template of the second user is associated with a predicted image of the second user after the time interval.

[0018] Clause 8: A method comprising: training, with at least one processor, a first machine learning model based on a training data set of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generating, with at least one processor, a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users within a time interval; generating, with at least one processor, a second machine learning model based on the plurality of image feature templates; generating, with at least one processor, a predicted image feature template using the second machine learning model; determining, with at least one processor, whether to authenticate an identity of a second user based on an input image of the second user, wherein determining whether to authenticate the identity of the second user based on the input image of the second user comprises generating a current image feature template based on the input image of the second user, comparing the current image feature template to the predicted image feature template, and determining whether the current image feature template corresponds to the predicted image feature template; and performing an action based on determining whether to authenticate the identity of the second user.

[0019] Clause 9: The method of clause 8, wherein training the first machine learning model based on the training data set of the plurality of images of the one or more first users comprises training the first machine learning model based on a training data set of a plurality of facial images of the one or more first users.

[0020] Clause 10: The method of clause 8 or 9, wherein generating the plurality of image feature templates using the first machine learning model comprises extracting a first image feature template from the first machine learning model after training the first machine learning model.

[0021] Clause 11: The method of any of clauses 8-10, further comprising: adding the input image of the second user to the plurality of images of the one or more first users in the training data set to provide an updated training data set; and retraining the first machine learning model based on the updated training data set.

[0022] Clause 12: The method of any of clauses 8-11, wherein generating the plurality of image feature templates using the first machine learning model comprises extracting a first image feature template from the first machine learning model after retraining the first machine learning model.

[0023] Clause 13: The method of any of clauses 8-12, wherein the first machine learning model is a convolutional neural network model and the second machine learning model is a long short-term memory recurrent neural network model.

[0024] Clause 14: The method of any of clauses 8-13, wherein the predicted image feature template of the second user is associated with a predicted image of the second user after the time interval.

[0025] Clause 15: A computer program product, the computer program product comprising at least one non-transitory computer-readable medium comprising one or more instructions, which, when executed by at least one processor, cause the at least one processor to: train a first machine learning model based on a training data set of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users within a time interval; generate a second machine learning model based on the plurality of image feature templates; generate a predicted image feature template using the second machine learning model; determine to authenticate an identity of a second user based on an input image of the second user, wherein the one or more instructions that cause the at least one processor to determine to authenticate the identity of the second user based on the input image of the second user cause the at least one processor to generate a current image feature template based on the input image of the second user, compare the current image feature template to the predicted image feature template, and determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on determining whether to authenticate the identity of the second user.

[0026] Clause 16: The computer program product of clause 15, wherein the one or more instructions that cause the at least one processor to train the first machine learning model cause the at least one processor to train the first machine learning model based on a training data set of a plurality of facial images of the one or more first users.

[0027] Clause 17: The computer program product of clause 15 or 16, wherein the one or more instructions that cause the at least one processor to generate the plurality of image feature templates using the first machine learning model cause the at least one processor to: extract a first image feature template from the first machine learning model after training the first machine learning model.

[0028] Clause 18: The computer program product of any one of clauses 15-17, wherein the one or more instructions further cause the at least one processor to: add the input image of the second user to the plurality of images of the one or more first users in the training data set to provide an updated training data set; and retrain the first machine learning model based on the updated training data set.

[0029] Clause 19: The computer program product of any one of clauses 15-18, wherein the one or more instructions that cause the at least one processor to generate the plurality of image feature templates using the first machine learning model cause the at least one processor to: extract a first image feature template from the first machine learning model after retraining the first machine learning model based on the updated training data set.

[0030] Clause 20: The computer program product of any one of clauses 15-19, wherein the predicted image feature template of the second user is associated with a predicted image of the second user after the time interval.

[0031] Clause 21: The computer program product of any one of clauses 15-20, wherein the first machine learning model is a convolutional neural network model and the second machine learning model is a long short-term memory recurrent neural network model.

[0032] These and other features and characteristics of the present disclosure, as well as the methods of operation and functions of the related elements of structures and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals designate similar, but not necessarily identical components. It is to be expressly understood, however, that the drawings are for purposes of illustration only and are not intended as a definition of the limits of the disclosure. As used in the specification and in the claims, the singular form of “a”, “an”, and “the” include plural referents unless the context clearly dictates otherwise. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1 is a diagram of a non-limiting embodiment or aspect of an environment in which the systems, apparatuses, products, devices, and / or methods described herein can be implemented in accordance with the principles of the disclosure;

[0034] Figure 2 is Figure 1 a diagram of non-limiting embodiments or aspects of one or more apparatuses and / or components of one or more systems in accordance with the present disclosure;

[0035] Figure 3 a flowchart of non-limiting embodiments or aspects of a process for authenticating an individual using image feature templates in accordance with the present disclosure; and

[0036] Figures 4A-4H a diagram of non-limiting embodiments or aspects of an implementation of a process for authenticating an individual using image feature templates in accordance with the present disclosure. DETAILED DESCRIPTION

[0037] For the purposes of description herein, the terms "end," "upper," "lower," "right," "left," "vertical," "horizontal," "top," "bottom," "lateral," "longitudinal," and derivatives thereof shall relate to the disclosure as oriented in the drawings. However, it is to be understood that the disclosure can assume various alternative orientations and step sequences, except where expressly specified to the contrary. It is also to be understood that the specific devices and processes illustrated in the attached drawings, and described in the following specification are simply exemplary embodiments or aspects of the present disclosure. Hence, specific dimensions and other physical characteristics related to the embodiments or aspects disclosed herein are not to be considered as limiting, unless otherwise indicated.

[0038] No aspect, component, element, structure, act, step, function, instruction, or the like as used herein should be construed as critical or essential unless explicitly described as such. Also, as used herein, the article "a" is intended to include one or more items, and can be used interchangeably with "one or more" and "at least one." Furthermore, as used herein, the term "set" is intended to include one or more items (e.g., related items, unrelated items, a combination of related and unrelated items, etc.), and can be used interchangeably with "one or more" or "at least one." Where only one item is intended, the term "one" or similar language is used. Also, as used herein, the

[0039] As used herein, the terms “communication” and “communicate” can refer to the reception, receipt, transmission, transfer, provision, and / or the like of information (e.g., data, signals, messages, instructions, commands, and / or the like). One unit (e.g., a device, a system, a component of a device or system, a combination thereof, and / or the like) being in communication with another unit means that the one unit is able to directly or indirectly receive

[0040] As used herein, the terms “issuer,” “issuer institution,” “issuer bank,” or “payment device issuer” can refer to one or more entities that provide accounts to individuals (e.g., users, customers, and / or the like) for conducting payment transactions, such as credit payment transactions and / or debit payment transactions. For example, an issuer institution can provide a customer with an account identifier, such as a primary account number (PAN), that uniquely identifies one or more accounts associated with the customer. In some non-limiting embodiments or aspects, an issuer can be associated with a bank identification number (BIN) that uniquely identifies the issuer institution. As used herein, an “issuer system” can refer to one or more computer systems operated by or on behalf of an issuer, such as a server executing one or more software applications. For example, an issuer system can include one or more authorization servers for authorizing transactions.

[0041] As used herein, the term “transaction service provider” can refer to an entity that receives transaction authorization requests from merchants or other entities and, in some cases, provides payment guarantees through an agreement between the transaction service provider and an issuer institution. For example, a transaction service provider can include a payment network, such as American or any other entity that processes transactions. As used herein, the term “transaction service provider system” can refer to one or more computer systems operated by or on behalf of a transaction service provider, e.g., a transaction service provider system executing one or more software applications. A transaction service provider system can include one or more processors and, in some non-limiting embodiments or aspects, can be operated by or on behalf of a transaction service provider.

[0042] As used herein, the term “merchant” can refer to one or more entities (e.g., an operator of a retail business) that provide goods and / or services and / or access to goods and / or services to users (e.g., customers, consumers, etc.) based on transactions, such as payment transactions. As used herein, a “merchant system” can refer to one or more computer systems operated by or on behalf of a merchant, e.g., a server executing one or more software applications. As used herein, the term “product” can refer to one or more goods and / or services provided by a merchant.

[0043] As used herein, the term “acquirer” can refer to an entity that is licensed by and approved by a transaction service provider to initiate transactions (e.g., payment transactions) involving payment devices associated with the transaction service provider. As used herein, the term “acquirer system” can also refer to one or more computer systems, computer devices, etc., operated by or on behalf of an acquirer. Transactions that an acquirer can initiate can include payment transactions (e.g., purchases, original credit transactions (OCTs), account funding transactions (AFTs), etc.). In some non-limiting embodiments or aspects, an acquirer can be authorized by a transaction service provider to contract with merchants or service providers to initiate transactions involving payment devices associated with the transaction service provider. An acquirer can contract with a payment facilitator to enable the payment facilitator to offer sponsorship to merchants. An acquirer can monitor the compliance of a payment facilitator according to transaction service provider regulations. An acquirer can conduct due diligence on a payment facilitator and ensure that proper due diligence occurs prior to contracting with a sponsored merchant. An acquirer can be liable for all transaction service provider programs operated or sponsored by the acquirer. An acquirer can be responsible for the actions of an acquirer payment facilitator, merchants sponsored by the acquirer payment facilitator, etc. In some non-limiting embodiments or aspects, an acquirer can be a financial institution, e.g., a bank.

[0044] As used herein, the term “payment gateway” can refer to an entity and / or a payment processing system operated by or on behalf of such an entity (e.g., a merchant service provider, a payment service provider, a payment facilitator, a payment facilitator under contract with an acquirer, a payment aggregator, etc.) that provides payment services (e.g., transaction service provider payment services, payment processing services, etc.) to one or more merchants. The payment services can be associated with the use of portable financial devices managed by a transaction service provider. As used herein, the term “payment gateway system” can refer to one or more computer systems, computer devices, servers, groups of servers, etc. operated by or on behalf of a payment gateway.

[0045] As used herein, the term “point of sale (POS) device” can refer to one or more devices that can be used by a merchant to conduct a transaction (e.g., a payment transaction) and / or process a transaction. For example, a POS device can include one or more client devices. Additionally or alternatively, a POS device can include a peripheral device, a card reader, a scanning device (e.g., a code scanner), a communication receiver, a near field communication (NFC) receiver, a radio frequency identification (RFID) receiver, and / or other contactless transceiver or receiver, a contact-based receiver, a payment terminal, etc.

[0046] As used herein, the term “point of sale (POS) system” can refer to one or more client devices and / or peripheral devices used by a merchant to conduct a transaction. For example, a POS system can include one or more POS devices, and / or other similar devices that can be used to conduct a payment transaction. In some non-limiting embodiments or aspects, a POS system (e.g., a merchant POS system) can include one or more server computers programmed or configured to process online payment transactions through a web page, a mobile application, etc.

[0047] As used herein, the terms “client” and “client device” can refer to one or more computing devices, such as a processor, a storage device, and / or similar computer components that access services that can be provided by a server. In some non-limiting embodiments or aspects, a client device can include an electronic device configured to communicate with one or more networks and / or facilitate a payment transaction, such as, but not limited to, one or more desktop computers, one or more portable computers (e.g., tablet computers), one or more mobile devices (e.g., cellular phones, smart phones, personal digital assistants, wearable devices such as watches, glasses, lenses, and / or clothing, etc.), and / or other similar devices. Further, the term “client” can also refer to an entity that owns, uses, and / or operates a client device to facilitate a transaction with another entity.

[0048] As used herein, the term "server" can refer to one or more computing devices, such as processors, storage devices, and / or similar computer components, that communicate with client devices and / or other computing devices over a network, such as the Internet or a private network, and in some instances, facilitate communication between other servers and / or client devices.

[0049] As used herein, the term "system" can refer to one or more computing devices or combinations of computing devices, such as but not limited to processors, servers, client devices, software applications, and / or other similar components. Moreover, as used herein, a reference to a "server" or a "processor" can refer to a previously described server and / or processor recited as performing a previous step or function, a different server and / or processor, and / or a combination of servers and / or processors. For example, as used in the specification and claims, a first server and / or a first processor recited as performing a first step or function can refer to the same or different server and / or processor recited as performing a second step or function.

[0050] Some embodiments or aspects are described herein in connection with thresholds. As used herein, satisfying a threshold can refer to a value being greater than the threshold, more than the threshold, higher than the threshold, greater than or equal to the threshold, less than the threshold, fewer than the threshold, lower than the threshold, less than or equal to the threshold, equal to the threshold, and / or the like.

[0051] Improved systems, methods, and computer program products for authenticating individuals using image feature templates are provided. Non-limiting embodiments or aspects of this disclosure may include a template authentication system comprising at least one processor programmed or configured to: train a first machine learning model (e.g., a feature template authentication machine learning model) based on a training dataset of multiple images of one or more first users, wherein the first machine learning model is configured to authenticate the identity of the one or more first users based on input images of the one or more first users; generate multiple image feature templates using the first machine learning model, wherein each of the multiple image feature templates is associated with positive authentication of the identity of the one or more first users within a time interval; generate a second machine learning model based on the multiple image feature templates; generate a predicted image feature template using the second machine learning model; determine the authentication of the second user based on an input image of a second user, wherein when the authentication of the second user is determined based on the input image of the second user, the at least one processor is programmed or configured to: generate a current image feature template based on the input image of the second user; compare the current image feature template with the predicted image feature template to determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on the determination of whether to authenticate the identity of the second user.

[0052] In this way, when aspects of an individual's physical characteristics change due to aging, non-limiting embodiments or aspects of this disclosure allow template authentication systems to accurately authenticate individuals based on images of their physical characteristics (e.g., facial images). Furthermore, non-limiting embodiments or aspects of this disclosure do not require obtaining and storing numerous samples of an individual's physical characteristics, thus reducing the requirements for technical resources (e.g., memory storage devices, processing resources, network resources, etc.).

[0053] Now for reference Figure 1 , Figure 1 This is an illustration of an exemplary environment 100 in which the apparatus, systems, methods, and / or products described herein can be implemented. Figure 1 As shown, environment 100 includes template authentication system 102, template authentication database 102a, and user device 104. Template authentication system 102, template authentication database 102a, and user device 104 can be interconnected via wired connection, wireless connection, or a combination of wired and wireless connection (e.g., establishing a communication connection).

[0054] The template authentication system 102 can include one or more computing devices configured to communicate with the template authentication database 102a and / or the user device 104 via the communication network 106. For example, the template authentication system 102 can include a server and / or other similar devices. In some non-limiting embodiments or aspects, the template authentication system 102 can be associated with a transaction service provider, as described herein. Additionally or alternatively, the template authentication system 102 can be associated with a merchant, a payment gateway, an acquirer institution, and / or an issuer system, as described herein.

[0055] The template authentication database 102a can include one or more computing devices configured to communicate with the template authentication system 102 and / or the user device 104 via the communication network 106. For example, the template authentication database 102a can include a server, a group of servers, and / or other similar devices. In some non-limiting embodiments or aspects, the template authentication database 102a can be associated with a transaction service provider, as described herein. Additionally or alternatively, the template authentication database 102a can be associated with a merchant, a payment gateway, an acquirer institution, and / or an issuer system, as described herein.

[0056] The user device 104 can include a computing device configured to communicate with the template authentication system 102 and / or the template authentication database 102a via the communication network 106. For example, the user device 104 can include a desktop computer, a portable computing machine (e.g., a laptop computer, a tablet computer, etc.), a mobile device (e.g., a cellular phone, a smart phone, a personal digital assistant, a wearable device such as a smart watch, smart glasses, etc.), and / or the like. In some non-limiting embodiments or aspects, the user device 104 can include an image capture device, such as a camera. The user device 104 can be configured to communicate via a short-range wireless communication connection (e.g., an NFC communication connection, an RFID communication connection, The user device 104 can transmit data to and / or receive data from the template authentication system 102 and / or the template authentication database 102a (e.g., via a data network, via a telephone network, via a power line communication network, via a computer network (e.g., the Internet), via a wireless communication connection, etc.). In some non-limiting embodiments or aspects, the user device 104 can be associated with a user (e.g., an individual operating the device). In some non-limiting embodiments or aspects, the user device 104 can include or can be a component of a device that enables a user to perform a financial transaction with an account provided by a financial institution. For example, the user device 104 can include or can be a component of an automated teller machine (ATM). In some non-limiting embodiments or aspects, the user device 104 can include or can be a component of a POS device or a POS system. For example, the user device 104 can include or can be a component of a POS device or a POS system of a self-checkout system (e.g., a non-cashier checkout system, a cashier-less checkout system, etc.). In some non-limiting embodiments or aspects, the user device 104 can be a component of the template authentication system 102.

[0057] The communication network 106 can include one or more wired and / or wireless networks. For example, the communication network 106 can include a cellular network (e.g., a long-term evolution (LTE) network, a third generation (3G) network, a fourth generation (4G) network, a code division multiple access (CDMA) network, etc.), a public land mobile network (PLMN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a telephone network (e.g., the public switched telephone network (PSTN)), a private network, an ad hoc network, an intranet, the Internet, a fiber optic-based network, a cloud computing network, and / or the like, and / or a combination of some or all of these or other types of networks.

[0058] Figure 1 The number and arrangement of systems and / or devices shown in FIG. 1 are provided as an example. There can be additional systems and / or devices, fewer systems and / or devices, different systems and / or devices, or differently arranged systems and / or devices than those shown in FIG. 1. Additionally or alternatively, a single system and / or a single device can implement one or more systems and / or devices shown in FIG. 1. Figure 1 The systems and / or devices shown in FIG. 1 can be implemented as any type of hardware unit, either alone or in combination with additional Figure 1 systems and / or devices. Additionally or alternatively, two or more Figure 1 systems and / or devices shown in FIG. 1 can be implemented within a single system and / or a single device.

[0059] Reference is now made to Figure 2 , Figure 2is a diagram of exemplary components of a device 200. The device 200 can correspond to one or more devices of the template authentication system 102, one or more devices of the template authentication database 102a, one or more devices of the user devices 104, and / or one or more devices of the communication network 106. In some non-limiting embodiments or aspects, one or more devices of the template authentication system 102, one or more devices of the template authentication database 102a, one or more devices of the user devices 104, and / or one or more devices of the communication network 106 can include at least one device 200 and / or at least one component of the device 200. As Figure 2 The device 200 can include a bus 202, a processor 204, a memory 206, a storage component 208, an input component 210, an output component 212, and a communication interface 214, as shown.

[0060] The bus 202 can include a component that permits communication among the components of the device 200. In some non-limiting embodiments or aspects, the processor 204 can be implemented in hardware, software, or a combination of hardware and software. For example, the processor 204 can include a processor (e.g., a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), etc.), a microprocessor, a digital signal processor (DSP), and / or any processing component that can be programmed to perform a function (e.g., a field programmable gate array (FPGA), an application- specific integrated circuit (ASIC), etc.). The memory 206 can include a random access memory (RAM), a read only memory (ROM), and / or another type of dynamic or static storage (e.g., flash memory, magnetic storage, optical storage, etc.) that stores information and / or instructions for use by the processor 204.

[0061] The storage component 208 can store information and / or software related to the operation and use of the device 200. For example, the storage component 208 can include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optic disk, a solid state disk, etc.), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a magnetic tape, and / or another type of computer- readable medium, along with a corresponding drive.

[0062] The input component 210 can include a component that permits the device 200 to receive information, such as via user input (e.g., a touch screen display, a keyboard, a keypad, a mouse, a button, a switch, a microphone, a camera, etc.). Additionally, or alternatively, the input component 210 can include a sensor (e.g., a global positioning system (GPS) component, an accelerometer, a gyroscope, an actuator, etc.) for sensing information. The output component 212 can include a component that provides output information from the device 200 (e.g., a display, a speaker, one or more light-emitting diodes (LEDs), etc.).

[0063] The communication interface 214 can include a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, etc.) that enables the device 200 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of both. The communication interface 214 can permit the device 200 to receive information from another device and / or provide information to another device. For example, the communication interface 214 can include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi® interface, a Bluetooth® interface, or the like. The communication interface 214 can include a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, etc.) that enables the device 200 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of both. The communication interface 214 can permit the device 200 to receive information from another device and / or provide information to another device. For example, the communication interface 214 can include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi® interface, a Bluetooth® interface, or the like. The communication interface 214 can include a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, etc.) that enables the device 200 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of both. The communication interface 214 can permit the device 200 to receive information from another device and / or provide information to another device. For example, the communication interface 214 can include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi® interface, a Bluetooth® interface, or the like. The communication interface 214 can include a transceiver-like component (e.g., a transceiver, a separate receiver and transmitter, etc.) that enables the device 200 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of both. The communication interface 214 can permit the device 200 to receive information from another device and / or provide information to another device. For example, the communication interface 214 can include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi® interface, a Bluetooth® interface, or the like.

[0064] The device 200 can perform one or more processes described herein. The device 200 can perform these processes based on processor 204 executing software instructions stored by a computer-readable medium, such as memory 206 and / or storage component 208. A computer-readable medium (e.g., a non-transitory computer-readable medium) is defined herein as a non-transitory memory device. A non-transitory memory device includes a memory space

[0065] The software instructions can be read into the memory 206 and / or the storage component 208 from another computer-readable medium or from another device via the communication interface 214. When executed, the software instructions stored in the memory 206 and / or the storage component 208 can cause the processor 204 to perform one or more processes described herein. Additionally, or alternatively, hardwired circuitry can be used in place of, or in combination with, software instructions to implement one or more processes described herein. Thus, embodiments or aspects described herein are not limited to any specific combination of hardware circuitry and software.

[0066] The memory 206 and / or the storage component 208 can include a data store or one or more data structures (e.g., a database, etc.). The device 200 can receive information from, store information in, transmit information to, or retrieve information from the data store or one or more data structures in the memory 206 and / or the storage component 208. For example, the information can include input data, output data, transaction data, account data, or any combination thereof.

[0067] Figure 2 The number and arrangement of components shown in FIG. 1 are provided as an example. In some non-limiting embodiments or aspects, device 200 can include additional components, fewer components, different components, or differently arranged components than those shown in FIG. 1. Figure 2The components are arranged in those different ways as shown. Alternatively, a group of components of device 200 (e.g., one or more components) may perform one or more functions described as being performed by another group of components of device 200.

[0068] Now for reference Figure 3 , Figure 3 This is a flowchart of some non-limiting embodiments or aspects of a process 300 for authenticating individuals using image feature templates. In some non-limiting embodiments or aspects, one or more functions described regarding process 300 may be performed by template authentication system 102 (e.g., fully, partially, etc.). In some non-limiting embodiments or aspects, one or more steps of process 300 may be performed by another device or set of devices (e.g., template authentication database 102a and / or user device 104) (e.g., fully, partially, etc.) that is separate from and / or includes said template authentication system 102.

[0069] like Figure 3 As shown, at step 302, process 300 may include generating multiple image feature templates. For example, template authentication system 102 may generate multiple image feature templates for a user associated with user device 104. In some non-limiting embodiments or aspects, the multiple image feature templates for a user may be associated with a time interval. For example, template authentication system 102 may generate multiple image feature templates for a user associated with user device 104 within that time interval. In another example, template authentication system 102 may generate each of the multiple image feature templates for a user based on one or more input images of the user received within that time interval. In some non-limiting embodiments or aspects, an image feature template may refer to a set of features computed from an image in a similar manner using a machine learning model. In some non-limiting embodiments or aspects, the image may be an image of the user's physical characteristics. For example, the image may include a facial image, such as an image of at least a portion of an individual's face, which may be used for the identification and / or authentication of the individual. In some non-limiting embodiments or aspects, the feature template may be an n-dimensional vector, where the dimensions of the vector include values ​​representing features of the image. In some non-limiting embodiments or aspects, the template authentication system 102 can generate one or more feature templates by extracting one or more feature templates from one or more layers (e.g., one or more layers) of a first machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate each of a plurality of image feature templates by extracting each image feature template from the first machine learning model. For example, the template authentication system 102 can generate each of a plurality of image feature templates by extracting each image feature template from the first machine learning model after training the first machine learning model.

[0070] In some non-limiting embodiments or aspects, the first machine learning model can be a convolutional neural network (CNN) model (e.g., a machine learning model that includes a CNN architecture). In this way, having the first machine learning model be a CNN model can require less memory to run machine learning operations and allow for training of larger, more powerful networks than other machine learning network models. In this way, the template authentication system 102 can generate more effective machine learning network models more quickly and / or with fewer resources than another type of machine learning network model (e.g., a multi-layer perceptron).

[0071] In some non-limiting embodiments or aspects, the template authentication system 102 can generate each of the plurality of image feature templates for a time point of the time interval. For example, the template authentication system 102 can generate a first image feature template for the user with respect to a first time point of the time interval, a second image feature template for the user with respect to a second time point of the time interval, a third image feature template for the user with respect to a third time point of the time interval, and an additional image feature template for the user with respect to an additional time point of the time interval, as appropriate.

[0072] In some non-limiting embodiments or aspects, the template authentication system 102 can generate each image feature template for each time point of the time interval based on the user’s input images provided by the user device 104 as input to the template authentication system 102 at each time point of the time interval (e.g., the user’s facial images provided by the user device 104 as input to the template authentication system 102 at the time point). For example, the template authentication system 102 can generate a first image feature template about a first time point based on a first input image of the user (e.g., a first input image of the user received at the first time point), a second image feature template about a second time point based on a second input image of the user (e.g., a second input image of the user received at the second time point), and a third image feature template about a third time point based on a third input image of the user (e.g., a third input image of the user received at the third time point). The first, second, and third input images can represent images of the user captured in a sequence. For example, the first time point of the time interval can be a time point before the second time point of the time interval, and the second time point of the time interval can be a time point before the third time point of the time interval. About each of the first, second, and third input images, each input image can be associated with a positive authentication of the identity of the user within the time interval. For example, each of the first, second, and third input images can have been determined by the template authentication system 102 to be a genuine image of the user within the time interval. In some non-limiting embodiments or aspects, the template authentication system 102 can generate each of the plurality of image feature templates at the respective time point of the time interval or after the time interval.

[0073] In some non-limiting embodiments or aspects, the template authentication system 102 can assign each image feature template for each time point of the time interval a timestamp of the respective time point. For example, the template authentication system 102 can assign the first image feature template a first timestamp based on the first time point, the second image feature template a second timestamp based on the second time point, and the third image feature template a third timestamp based on the third time point.

[0074] In some non-limiting embodiments or aspects, the template authentication system 102 can generate a first machine learning model, where the first machine learning model is configured to authenticate an identity of a user (e.g., an individual) based on an image of the user. In one example, the first machine learning model can be configured to provide a prediction about whether an image of the individual is a genuine image of the user (e.g., an image of the user determined to be genuine by the template authentication system 102), and the prediction can be used (e.g., by the template authentication system 102 or another authentication system) to authenticate the identity of the user.

[0075] In some non-limiting embodiments or aspects, the template authentication system 102 can train (e.g., initially train) the first machine learning model based on a training dataset that includes a plurality of images of a user (e.g., a plurality of facial images of the user). In some non-limiting embodiments or aspects, the training dataset can include a plurality of images of the user, where one or more images (e.g., all images, a set of images, one image, etc.) of the plurality of images are authentic images of the user. In some non-limiting embodiments or aspects, the template authentication system 102 can train the first machine learning model based on a training dataset that includes a plurality of images of one or more users, where the first machine learning model is configured to authenticate an identity of each user of the one or more users based on an input image of the user. In some non-limiting embodiments or aspects, the template authentication system 102 can store the plurality of image feature templates and / or the training dataset in the template authentication database 102a.

[0076] In some non-limiting embodiments or aspects, the template authentication system 102 can train the first machine learning model based on a loss function. For example, the template authentication system 102 can calculate a result of the loss function based on training the first machine learning model using each image of the plurality of images of the user in the training dataset. The template authentication system 102 can backpropagate the result of the loss function to update one or more weights of the first machine learning model.

[0077] In some non-limiting embodiments or aspects, the template authentication system 102 can authenticate an identity of a user using the first machine learning model. For example, the template authentication system 102 can receive an input image of the user from the user device 104 (e.g., an input image of the user captured using an image capture device, such as a camera, of the user device 104), and the template authentication system 102 can provide the input image of the user as an input to the first machine learning model. The template authentication system 102 can receive an output from the first machine learning model, the output including a prediction as to whether the image of the user is an authentic image of the user. The template authentication system 102 can determine to authenticate the identity of the user based on the output of the first machine learning model indicating that a positive authentication of the identity of the user has been made.

[0078] In some non-limiting embodiments or aspects, the template authentication system 102 can generate a plurality of image feature templates over a time interval using the first machine learning model. For example, the template authentication system 102 can generate a first image feature template of the plurality of image feature templates using the first machine learning model with respect to a first time point of the time interval, generate a second image feature template of the plurality of image feature templates using the first machine learning model with respect to a second time point of the time interval, generate a third image feature template of the plurality of image feature templates using the first machine learning model with respect to a third time point of the time interval, and generate an additional image feature template of the plurality of image feature templates using the first machine learning model with respect to an additional time point of the time interval. In some non-limiting embodiments or aspects, each image feature template of the plurality of image feature templates can be associated with a positive authentication of the identity of the user over the time interval. In one example, the template authentication system 102 can generate a first image feature template for the user using the first machine learning model after a first positive authentication of the identity of the user has been made (e.g., over the time interval), generate a second image feature template for the user using the first machine learning model after a second positive authentication of the identity of the user has been made, generate a third image feature template for the user using the first machine learning model after a third positive authentication of the identity of the user has been made, and generate an additional image feature template for the user using the first machine learning model after an additional positive authentication of the identity of the user has been made.

[0079] In some non-limiting embodiments or aspects, the template authentication system 102 can retrain the first machine learning model. For example, the template authentication system 102 can retrain the first machine learning model after positively authenticating the identity of the user. In such examples, the template authentication system 102 can determine, using the first machine learning model, whether to authenticate the identity of the user based on an input image of the user. The template authentication system 102 can determine to authenticate the identity of the user based on the input image of the user indicating that the identity of the user has been positively authenticated. The template authentication system 102 can add the input image of the user to the plurality of images in the training dataset (e.g., the training dataset from which the machine learning model was initially trained) to provide an updated training dataset. The template authentication system 102 can retrain the first machine learning model based on the updated training dataset (e.g., based on one or more images included in the updated training dataset). In some non-limiting embodiments or aspects, the template authentication system 102 can store the updated training dataset in the template authentication database 102a. In some non-limiting embodiments or aspects, the template authentication system 102 can generate one or more of the plurality of image feature templates by extracting one or more image feature templates from the first machine learning model after retraining the first machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can combine all input images of the user associated with the positive authentication of the identity of the user after a time interval to provide a new training dataset. In some non-limiting embodiments or aspects, the template authentication system 102 can retrain the first machine learning model based on one or more images included in the new training dataset. For example, the template authentication system 102 can retrain the first machine learning model based on receiving one input image of the user associated with the positive authentication of the identity of the user. In another example, the template authentication system 102 can retrain the first machine learning model based on receiving a set of input images (e.g., a batch of input images) of the user associated with the positive authentication of the identity of the user.

[0080] In some non-limiting embodiments or aspects, the template authentication system 102 can generate additional input images of the user. For example, the template authentication system 102 can generate the additional input images of the user using a generative adversarial network (GAN) model. In some non-limiting embodiments or aspects, the template authentication system 102 can add the additional input images of the user to the plurality of images in the training dataset (e.g., the training dataset from which the machine learning model was initially trained, the updated training dataset, the new training dataset, etc.).

[0081] As Figure 3As shown, at step 304, process 300 can include generating a predicted image feature template for the user. For example, template authentication system 102 can generate one or more predicted image feature templates for a user associated with user device 104. In some non-limiting embodiments or aspects, the user associated with user device 104 (e.g., a first user) can be different from a user (e.g., a second user) associated with the plurality of images included in the training dataset used to train the first machine learning model. In some non-limiting embodiments or aspects, the user associated with user device 104 can be the same as the user associated with the plurality of images included in the training dataset used to train the first machine learning model.

[0082] In some non-limiting embodiments or aspects, the predicted image feature template for the user can include an image feature template based on the user’s predicted image for a future time (e.g., at a certain point in time or time interval). The user’s predicted image for the future time can include a prediction of how the user will look during a future certain point in time or time interval. In some non-limiting embodiments or aspects, the predicted image feature template for the user can be associated with a predicted image of the user after a time interval (e.g., a historical time interval). For example, the predicted image feature for the user can be associated with a predicted image of the user after a time interval during which the identity of the user was positively authenticated one or more times by template authentication system 102.

[0083] In some non-limiting embodiments or aspects, template authentication system 102 can generate a predicted image feature template for the user based on the plurality of image feature templates. For example, template authentication system 102 can extract the predicted image feature template from the second machine learning model after training the second machine learning model using the plurality of image feature templates. In some non-limiting embodiments or aspects, template authentication system 102 can generate a plurality of predicted image feature templates for the user based on the plurality of image feature templates. In some non-limiting embodiments or aspects, template authentication system 102 can assign a timestamp to each predicted image feature template of the plurality of predicted image feature templates for the user based on a future time during which the respective predicted image feature template can be used to authenticate the identity of the user.

[0084] In some non-limiting embodiments or aspects, template authentication system 102 can generate a second machine learning model (e.g., a predicted image feature template generation machine learning model), where the second machine learning model is configured to generate a predicted image feature template for an individual (e.g., a user). For example, template authentication system 102 can generate the second machine learning model based on the plurality of image feature templates.

[0085] In some non-limiting embodiments or aspects, the second machine learning model can be a long short-term memory (LSTM) model (e.g., a machine learning model that includes an LSTM recurrent neural network architecture). In this way, the LSTM model can store historical information in memory during machine learning operations and allow for more accurate generation of predicted image feature templates as compared to other machine learning network models. In this way, the template authentication system 102 can generate predicted image feature templates more accurately for users with fewer resources as compared to other types of machine learning network models, such as multilayer perceptron or convolutional neural networks.

[0086] In some non-limiting embodiments or aspects, the template authentication system 102 can generate one or more predicted image feature templates using the second machine learning model. For example, the template authentication system 102 can generate one or more predicted image feature templates by extracting one or more predicted image feature templates from one or more layers (e.g., one or more layers) of the second machine learning model.

[0087] In some non-limiting embodiments or aspects, the template authentication system 102 can train (e.g., initially train) the second machine learning model based on a training dataset that includes a plurality of image feature templates. In some non-limiting embodiments or aspects, the training dataset can include a plurality of image feature templates, where one or more image feature templates (e.g., all image feature templates, a set of image feature templates, one image feature template, etc.) of a plurality of images are associated with positive authentication of an identity of a user. In some non-limiting embodiments or aspects, the template authentication system 102 can store the predicted image feature templates and / or the training dataset in the template authentication database 102a.

[0088] As Figure 3As shown in FIG. 3, at step 306, process 300 can include determining whether to authenticate the user based on the input image of the user. For example, template authentication system 102 can determine whether to authenticate the identity of the user associated with user device 104 during a runtime (e.g., real-time) process based on the input image of the user and the predicted image feature template of the user. In some non-limiting embodiments or aspects, template authentication system 102 can generate a current image feature template for the user based on the input image of the user (e.g., an image feature template generated during a runtime process), compare the current image feature template of the user to the predicted image feature template of the user, and determine whether the current image feature template of the user corresponds to the predicted image feature template of the user. If template authentication system 102 determines that the current image feature template of the user corresponds to the predicted image feature template of the user, template authentication system 102 can determine to authenticate the user. If template authentication system 102 determines that the current image feature template of the user does not correspond to the predicted image feature template of the user, template authentication system 102 can determine not to authenticate the user. In some non-limiting embodiments or aspects, template authentication system 102 can generate the current image feature template for the user using the first machine learning model. For example, template authentication system 102 can generate the current image feature template by extracting the current image feature template from one or more layers (e.g., one or more layers) of the first machine learning model. For example, template authentication system 102 can generate the current image feature template by extracting the current image feature template from the first machine learning model after providing the input image of the user as an input to the first machine learning model and / or after training the first machine learning model.

[0089] In some non-limiting embodiments or aspects, the template authentication system 102 can determine whether the current image feature template of the user corresponds to the predicted image feature template of the user based on a time associated with the runtime process. For example, the template authentication system 102 can determine a time associated with the runtime process (e.g., a point in time at which the runtime process is occurring or a time interval during which the runtime process is occurring). The template authentication system 102 can retrieve the predicted image feature template of the user from the plurality of predicted image feature templates of the user stored in the template authentication database 102a based on the time associated with the runtime process. In some non-limiting embodiments or aspects, the template authentication system 102 can retrieve the predicted image feature template of the user using the time associated with the runtime process. For example, the template authentication system 102 can select the predicted image feature template of the user from the plurality of predicted image feature templates of the user that is assigned a timestamp corresponding to the time associated with the runtime process. The template authentication system 102 can retrieve the predicted image feature template of the user, and the template authentication system 102 can compare the current image feature template of the user to the predicted image feature template of the user. The template authentication system 102 can determine whether the current image feature template of the user corresponds to the predicted image feature template of the user based on comparing the current image feature template of the user to the predicted image feature template of the user.

[0090] In some non-limiting embodiments or aspects, the template authentication system 102 can determine whether the current image feature template of the user corresponds to the predicted image feature template of the user based on a distance. For example, the template authentication system 102 can determine a distance (e.g., a Euclidean distance) between one or more values of the current image feature template of the user and one or more values of the predicted image feature template of the user. The template authentication system 102 can determine whether the distance satisfies a threshold by comparing the distance to the threshold. If the template authentication system 102 determines that the distance satisfies the threshold, the template authentication system 102 can determine that the current image feature template of the user corresponds to the predicted image feature template of the user. If the template authentication system 102 determines that the distance does not satisfy the threshold, the template authentication system 102 can determine that the current image feature template of the user does not correspond to the predicted image feature template of the user.

[0091] As Figure 3At step 308, process 300 can include performing an action based on determining whether to authenticate the user, as shown. For example, template authentication system 102 can perform an action based on determining whether to authenticate the identity of the user associated with user device 104. In some non-limiting embodiments or aspects, template authentication system 102 can perform an action associated with allowing or preventing access (e.g., accessing an account of the user, accessing a computer system, etc.) based on determining whether to authenticate the identity of the user associated with user device 104. For example, template authentication system 102 can perform an action associated with allowing access based on determining to authenticate the identity of the user associated with user device 104. In another example, template authentication system 102 can perform an action associated with preventing access based on determining not to authenticate the identity of the user associated with user device 104.

[0092] In some non-limiting embodiments or aspects, template authentication system 102 can perform an action associated with authorizing an operation (e.g., an operation associated with a runtime process, such as a payment transaction, an operation of a computer system to execute a runtime command, etc.) that is to be performed based on determining whether to authenticate the identity of the user associated with user device 104. For example, template authentication system 102 can perform an action associated with authorizing performing an operation based on determining to authenticate the identity of the user associated with user device 104. In another example, template authentication system 102 can perform an action associated with not authorizing performing an operation based on determining not to authenticate the identity of the user associated with user device 104.

[0093] In some non-limiting embodiments or aspects, template authentication system 102 can send a message regarding the authentication process. For example, template authentication system 102 can send a message to user device 104 that includes an indication that the identity of the user has been authenticated based on template authentication system 102 determining to authenticate the identity of the user. In another example, template authentication system 102 can send a message to user device 104 that includes an indication that the identity of the user has not been authenticated based on template authentication system 102 determining not to authenticate the identity of the user.

[0094] In some non-limiting embodiments or aspects, template authentication system 102 can send a message that includes a request for an additional authentication parameter to be provided by the user associated with user device 104 based on template authentication system 102 determining not to authenticate the identity of the user. For example, template authentication system 102 can send a message via user device 104 (e.g., via an image capture device of user device 104) that includes a request for the user to provide a gesture, such as a motion associated with waving a hand to confirm.

[0095] Reference is now made to Figure 4A - 4H, Figures 4A-4HThis is an illustration of implementation 400 of the process of authenticating an individual using image feature templates (e.g., process 300). Figure 4A As indicated by reference numeral 405 in the accompanying drawings, the template authentication system 102 can initially train a first machine learning model (e.g., a feature template authentication machine learning model). In some non-limiting embodiments or aspects, the template authentication system 102 can initially train the first machine learning model based on a training dataset including multiple facial images of the user. In some non-limiting embodiments or aspects, the training dataset can include multiple facial images of the user, wherein each of the multiple facial images is a real facial image of the user. In some non-limiting embodiments or aspects, the first machine learning model is configured to authenticate the user's identity based on an input facial image of the user (e.g., an input including the user's facial image provided to the template authentication system 102 by the user device 104). In some non-limiting embodiments or aspects, the first machine learning model can be a convolutional neural network (CNN) model (e.g., a machine learning model including a CNN architecture).

[0096] like Figure 4B As indicated by reference numeral 410 in the accompanying drawings, the template authentication system 102 can generate multiple image feature templates for a user associated with the user device 104 using a first machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate multiple image feature templates within a time interval using the first machine learning model. For example, the template authentication system 102 can generate a first image feature template among multiple image feature templates using the first machine learning model at a first point in time with respect to the time interval, a second image feature template among multiple image feature templates using the first machine learning model at a second point in time with respect to the time interval, and a third image feature template among multiple image feature templates using the first machine learning model at a third point in time with respect to the time interval. In some non-limiting embodiments or aspects, the time interval can be a time interval during which the user's identity has been affirmatively authenticated a predetermined number of times.

[0097] like Figure 4C As indicated by reference numeral 415 in the accompanying drawings, the template authentication system 102 can generate a second machine learning model (e.g., a predictive image feature template generation machine learning model) based on multiple image feature templates of a user associated with user device 104. In some non-limiting embodiments or aspects, the second machine learning model is configured to generate predicted image feature templates of the user associated with user device 104. In some non-limiting embodiments or aspects, the template authentication system 102 can initially train the second machine learning model based on a training dataset including multiple image feature templates.

[0098] like Figure 4DAs shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model.

[0099] As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. Figure 4E As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model.

[0100] As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. Figure 4F As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. 4G As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. Figure 4F As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. Figure 4G As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. Figure 4G As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model.

[0101] As shown in FIG. 4, the template authentication system 102 can generate a predicted image feature template for a user associated with the user device 104 using a second machine learning model. In some non-limiting embodiments or aspects, the template authentication system 102 can generate the predicted image feature template by extracting the predicted image feature template from one or more layers (e.g., one or more layers) of the second machine learning model. Figure 4HAs shown in FIG. 1, the template authentication system 102 can perform an action based on determining whether to authenticate the identity of the user associated with the user device 104. As Figure 4H As shown by reference number 445 in FIG. 4B, the template authentication system 102 can send a message to the user device 104 regarding the authentication process. For example, the template authentication system 102 can send a message that includes an indication that the identity of the user has been authenticated based on the template authentication system 102 determining to authenticate the identity of the user. In another example, the template authentication system 102 can send a message that includes an indication that the identity of the user has not been authenticated based on the template authentication system 102 determining not to authenticate the identity of the user.

[0102] In some non-limiting embodiments or aspects, the template authentication system 102 can send a message that includes a request for the user associated with the user device 104 to provide an additional authentication parameter based on the template authentication system 102 determining not to authenticate the identity of the user. For example, the template authentication system 102 can send a message via the user device 104 (e.g., via an image capture device of the user device 104) that includes a request for the user to provide a gesture, such as a motion associated with waving a hand to confirm.

[0103] While the above method, system and computer program product have been described in detail for the purpose of illustration based on what is currently considered to be the most practical and preferred embodiments or aspects, it is understood that such detail is solely for that purpose and that the disclosure is not limited to the described embodiments or aspects, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims. For instance, it is to be understood that the present disclosure contemplates that, to the extent possible, one or more features of any embodiment or aspect can be combined with one or more features of any other embodiment or aspect.

Claims

1. A system comprising: at least one processor programmed or configured to: train a first machine learning model based on a training dataset of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users during a time interval, wherein each image feature template is a multi-dimensional vector, wherein dimensions of the vector comprise values representing features of an image, and wherein, when generating the plurality of image feature templates, the at least one processor is programmed or configured to: generate each image feature template of the plurality of image feature templates for a point in time of the time interval for the one or more first users based on one or more input images of the one or more first users received during the time interval that resulted in a positive authentication of the identity of the one or more first users during the time interval; generate a second machine learning model based on the plurality of image feature templates; generate a predicted image feature template using the second machine learning model, wherein the predicted image feature template comprises an image feature template based on a predicted image of a second user with respect to a future time after the time interval; determine to authenticate an identity of the second user based on an input image of the second user received after the time interval, wherein, when determining to authenticate the identity of the second user based on the input image of the second user, the at least one processor is programmed or configured to: generate a current image feature template based on the input image of the second user, compare the current image feature template to the predicted image feature template, and determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on determining to authenticate the identity of the second user.

2. The system of claim 1, wherein, when training the first machine learning model based on the training dataset of the plurality of images of the one or more first users, the at least one processor is programmed or configured to: train the first machine learning model based on a training dataset of a plurality of facial images of the one or more first users.

3. The system of claim 1, wherein, when generating the plurality of image feature templates using the first machine learning model, the at least one processor is programmed or configured to: extract a first image feature template from the first machine learning model after training the first machine learning model.

4. The system of claim 1, wherein, the at least one processor is programmed or configured to: add the input image of the second user to the plurality of images of the one or more first users in the training dataset to provide an updated training dataset; and retrain the first machine learning model based on the updated training dataset. when generating the plurality of image feature templates using the first machine learning model, the at least one processor is programmed or configured to:

5. The system of claim 4, wherein, ​ extracting, from the first machine learning model, a first image feature template after retraining the first machine learning model based on the updated training dataset of images.

6. The system of claim 1, wherein, the first machine learning model is a convolutional neural network model and the second machine learning model is a long short-term memory recurrent neural network model.

7. The system of claim 1, wherein, the predicted image feature template of the second user is associated with a predicted image of the second user after the time interval.

8. A method comprising: training, with at least one processor, a first machine learning model based on a training dataset of a plurality of images of one or more first users, wherein the first machine learning model is configured to authenticate an identity of the one or more first users based on an input image of the one or more first users; generating, with at least one processor, a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users within a time interval, wherein each image feature template is a multi-dimensional vector, wherein dimensions of the vector comprise values representing features of an image, and wherein, when generating the plurality of image feature templates, the at least one processor is programmed or configured to: generate, for a point in time of the time interval, each image feature template of the plurality of image feature templates for the one or more first users based on one or more input images of the one or more first users received during the time interval that resulted in a positive authentication of the identity of the one or more first users during the time interval; generating, with at least one processor, a second machine learning model based on the plurality of image feature templates; generating, with at least one processor, a predicted image feature template using the second machine learning model, wherein the predicted image feature template comprises an image feature template based on a predicted image of a second user with respect to a future time after the time interval; determining, with at least one processor, whether to authenticate an identity of the second user based on an input image of the second user received after the time interval, wherein determining whether to authenticate the identity of the second user based on the input image of the second user comprises: generating a current image feature template based on the input image of the second user; comparing the current image feature template to the predicted image feature template; and determining whether the current image feature template corresponds to the predicted image feature template; and performing an action based on determining whether to authenticate the identity of the second user.

9. The method of claim 8, wherein, training the first machine learning model based on the training dataset of a plurality of images of the one or more first users comprises: training the first machine learning model based on a training dataset of a plurality of facial images of the one or more first users.

10. The method of claim 8, wherein, generating the plurality of image feature templates using the first machine learning model comprises: extracting, from the first machine learning model, a first image feature template after training the first machine learning model.

11. The method of claim 8, further comprising: adding the input image of the second user to the plurality of images of the one or more first users in the training dataset to provide an updated training dataset; and retraining the first machine learning model based on the updated training dataset.

12. The method of claim 11, wherein, generating the plurality of image feature templates using the first machine learning model includes: extracting a first image feature template from the first machine learning model after retraining the first machine learning model.

13. The method of claim 8, wherein, the first machine learning model is a convolutional neural network model and the second machine learning model is a long short-term memory recurrent neural network model.

14. The method of claim 8, wherein, the predicted image feature template of the second user is associated with a predicted image of the second user after the time interval.

15. A computer program product, the computer program product comprising at least one non-transitory computer-readable medium comprising one or more instructions configured to, when executed, cause at least one processor to: training a first machine learning model based on a training data set of a plurality of images of one or more first users, wherein, the first machine learning model is configured to authenticate an identity of the one or more first users based on input images of the one or more first users; generate a plurality of image feature templates using the first machine learning model, wherein each image feature template of the plurality of image feature templates is associated with a positive authentication of the identity of the one or more first users within a time interval, wherein each image feature template is a multi-dimensional vector, wherein dimensions of the vector comprise values representing features of an image, and wherein the one or more instructions that cause the at least one processor to generate the plurality of image feature templates cause the at least one processor to: generate each image feature template of the plurality of image feature templates for a time point of the time interval for the one or more first users based on one or more input images received by the user during the time interval that resulted in a positive authentication of the identity of the one or more first users during the time interval; generate a second machine learning model based on the plurality of image feature templates; generate a predicted image feature template using the second machine learning model, wherein the predicted image feature template comprises an image feature template based on a predicted image of a second user with respect to a future time after the time interval; determine to authenticate an identity of the second user based on an input image of the second user received after the time interval, wherein the one or more instructions that cause the at least one processor to determine to authenticate an identity of the second user based on an input image of the second user cause the at least one processor to: generate a current image feature template based on the input image of the second user, compare the current image feature template to the predicted image feature template, and determine that the current image feature template corresponds to the predicted image feature template; and perform an action based on determining whether to authenticate the identity of the second user.

16. The computer program product of claim 15, wherein, the one or more instructions that cause the at least one processor to train the first machine learning model cause the at least one processor to: training the first machine learning model based on a training dataset of a plurality of facial images of the one or more first users.

17. The computer program product of claim 15, wherein, The one or more instructions that cause the at least one processor to generate the plurality of image feature templates using the first machine learning model cause the at least one processor to: extract a first image feature template from the first machine learning model after training the first machine learning model.

18. The computer program product of claim 15, wherein the one or more instructions further cause the at least one processor to: add the input image of the second user to the plurality of images of the one or more first users in the training dataset to provide an updated training dataset; and retrain the first machine learning model based on the updated training dataset.

19. The computer program product of claim 18, wherein, The one or more instructions that cause the at least one processor to generate the plurality of image feature templates using the first machine learning model cause the at least one processor to: extract a first image feature template from the first machine learning model after training the first machine learning model.

20. The computer program product of claim 15, wherein, The predicted image feature template of the second user is associated with a predicted image of the second user after the time interval. The one or more instructions that cause the at least one processor to generate the plurality of image feature templates using the first machine learning model cause the at least one processor to: extract a first image feature template from the first machine learning model after training the first machine learning model. The predicted image feature template of the second user is associated with a predicted image of the second user after the time interval.

Images