Age information generation method and apparatus, electronic device, and computer-readable medium

Abstract

Embodiments of the present disclosure disclose an age information generation method, device, electronic equipment and computer readable medium. A specific implementation of the method comprises: obtaining a face image of a target living body to be age-identified; inputting the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image; inputting the face image into a first feature extraction network to generate a first feature map; generating a second feature map set representing local feature information of each face key region according to the first feature map and the probability map corresponding to each face key region; and generating age information of the target living body according to the second feature map set and the first feature map. The implementation can generate more accurate age information of the target living body.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of computer, and particularly, to an age information generation method and device, an electronic device and a computer readable medium. BACKGROUND

[0002] At present, with the rapid development of biological face recognition technology, the corresponding biological age information recognition is also widely used in various recognition scenarios. For example, through a face image, whether the corresponding human body is a minor is identified. For the generation of biological age information, the commonly used way is to directly input the obtained image including the face of the biological body into a pre-trained convolutional neural network (CNN) to output the corresponding age information.

[0003] However, when the above method is used to determine the biological age information, the following technical problems often exist:

[0004] The face feature information extracted by the convolutional neural network is often the overall face feature information in the image, and the generation of the age information only according to the overall face feature information often has the problem of low accuracy. SUMMARY

[0005] The summary part of the present disclosure is used to introduce the concepts in a brief form, which will be described in detail in the specific embodiments part. The summary part of the present disclosure is not intended to identify the key features or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of the present disclosure propose an age information generation method, device, electronic device and computer readable medium to solve the technical problems mentioned in the background part.

[0007] In a first aspect, some embodiments of the present disclosure provide an age information generation method, comprising: obtaining a face image of a target biological body to be age-identified; inputting the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image; inputting the face image into a first feature extraction network to generate a first feature map; generating a second feature map set representing local feature information of the each face key region according to the first feature map and the probability map corresponding to the each face key region; and generating age information of the target biological body according to the second feature map set and the first feature map.

[0008] In a second aspect, some embodiments of the present disclosure provide an age information generation apparatus, comprising: an acquisition unit configured to acquire a face image of a target living body to be age-identified; a first input unit configured to input the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image; a second input unit configured to input the face image into a first feature extraction network to generate a first feature map; a first generation unit configured to generate a second feature map set representing local feature information of the each face key region according to the first feature map and the probability map corresponding to the each face key region; and a second generation unit configured to generate age information of the target living body according to the second feature map set and the first feature map.

[0009] In a third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; and a storage device having one or more programs stored thereon, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method described in any implementation manner of the first aspect.

[0010] In a fourth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0011] The above various embodiments of the present disclosure have the following beneficial effects: the age information generation method of some embodiments of the present disclosure can accurately generate age information of a target organism. Specifically, the reason why the age information of the target organism cannot be accurately generated is that only the overall facial feature information of the face image to be recognized is used in the process of generating the age information of the target organism, resulting in inaccurate generation of the age information of the target organism. Based on this, the age information generation method of some embodiments of the present disclosure can first acquire a face image of a target organism to be age-identified for subsequent determination of the age information of the target organism. Then, by inputting the face image into a face semantic segmentation network, a probability map corresponding to each facial key region can be accurately and efficiently generated. Next, the face image is input into a first feature extraction network to generate a first feature map that can represent overall facial feature information. Furthermore, according to the first feature map and the probability map corresponding to each facial key region, a second feature map set representing local feature information of the facial key regions is accurately generated. Here, the generated second feature map set is used for subsequent generation of the age information of the target organism. Finally, for the second feature map set and the first feature map, on the basis of multi-level feature information, not only the overall facial feature information of the face image is considered, but also the local facial feature information is considered, so that the subsequently generated age information is more accurate. Thus, the age information generation method can accurately generate the age information of the target organism. BRIEF DESCRIPTION OF DRAWINGS

[0012] The above and other features, aspects and advantages of the embodiments of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and elements and features are not necessarily to scale.

[0013] Figure 1 is a schematic diagram of an application scenario of an age information generation method according to some embodiments of the present disclosure;

[0014] Figure 2 is a flowchart of some embodiments of the age information generation method according to the present disclosure;

[0015] Figure 3 is a flowchart of another embodiment of the age information generation method according to the present disclosure;

[0016] Figure 4 is an example diagram of a network structure of a first feature extraction network in some embodiments of the age information generation method according to the present disclosure;

[0017] Figure 5is a structural schematic diagram of some embodiments of an age information generation apparatus according to the present disclosure;

[0018] Figure 6 is a structural schematic diagram of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0019] Embodiments of the present disclosure will be described below in greater detail with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be interpreted as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly and completely understood. It should be understood that the drawings and embodiments of the present disclosure are only for illustrative purposes and are not intended to limit the scope of protection of the present disclosure.

[0020] It should also be noted that, for ease of description, only parts related to the present application are shown in the drawings. The embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0021] It should be noted that the terms "first", "second", etc. mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0022] It should be noted that the adjectives "one", "multiple" mentioned in the present disclosure are illustrative and not restrictive, and those skilled in the art should understand that, unless otherwise explicitly stated in the context, it should be understood as "one or more".

[0023] The names of the messages or information exchanged between the devices in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0024] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0025] Figure 1 is a structural schematic diagram of some embodiments of an age information generation apparatus according to the present disclosure;

[0026] In Figure 1In this application scenario, the electronic device 101 can first acquire a facial image 102 of the target organism to be identified in terms of age. Then, the electronic device 101 can input the facial image 102 into a pre-trained facial semantic segmentation network 103 to output probability maps 104 corresponding to each key facial region in the facial image 102. Optionally, the probability maps 104 corresponding to each key facial region may include a probability map corresponding to the mouth region and a probability map corresponding to the eye region. Next, the electronic device 101 can input the facial image 102 into a first feature extraction network 105 to generate a first feature map 106. Furthermore, the electronic device 101 can generate a second feature map set 107 representing the local feature information of each key facial region based on the first feature map 106 and the probability maps 104 corresponding to each key facial region. Finally, the electronic device 101 can generate the age information 108 of the target organism based on the second feature map set 107 and the first feature map 106. In this application scenario, the age information 108 can be "44".

[0027] It should be noted that the aforementioned electronic device 101 can be either hardware or software. When the electronic device is hardware, it can be implemented as a distributed cluster consisting of multiple server or terminal devices, or as a single server or terminal device. When the electronic device is software, it can be installed in the hardware devices listed above. It can be implemented as, for example, multiple software programs or software modules used to provide distributed services, or as a single software program or software module. No specific limitations are made here.

[0028] It should be understood that Figure 1 The number of electronic devices shown is merely illustrative. Any number of electronic devices can be used depending on the implementation requirements.

[0029] Continue to refer to Figure 2 The diagram illustrates a flow 200 of some embodiments of an age information generation method according to the present disclosure. This age information generation method includes the following steps:

[0030] Step 201: Obtain the facial image of the target organism to be identified in terms of age.

[0031] In some embodiments, the entity executing the above-described age information generation method (e.g.) Figure 1 The electronic device 101 shown can acquire a facial image of a target organism to be identified in terms of age via a wired or wireless connection. The target organism can be a pre-determined organism among at least one organism displayed in the facial image.

[0032] Optionally, the face image can be a pre-processed image. For example, the face image can be a face pose tilt-corrected image.

[0033] At step 202, the face image is input into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image.

[0034] In some embodiments, the execution subject can input the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image. The face semantic segmentation network can be a network for performing semantic segmentation on a face image. The elements in the probability map corresponding to the face key region with a probability greater than or equal to a predetermined value are elements corresponding to the face key region. The face semantic segmentation network can be, but is not limited to, at least one of the following: a U-net network, a Mask RCNN (Regions with CNN features) network based on a residual network.

[0035] In some optional implementations of some embodiments, the face image is a human face image, and the face key regions include an eye region, a mouth region, a nose region, and a forehead region. The eye key region can be a region containing the eyes of the target biological body. The mouth key region can be a region containing the mouth of the target biological body. The nose key region can be a region containing the nose of the target biological body. The forehead key region can be a region containing the forehead of the target biological body.

[0036] At step 203, the face image is input into a first feature extraction network to generate a first feature map.

[0037] In some embodiments, the execution subject can input the face image into a first feature extraction network to generate a first feature map. The first feature extraction network can be a network for extracting image feature information in the face image. The first feature map can represent the global feature information of the face image.

[0038] As an example, the execution subject can input the face image into a multi-layer serially connected residual network (Residual Networks, ResNets) to generate the first feature map.

[0039] In some optional implementations of some embodiments, the first feature extraction network includes a plurality of serially connected second convolutional networks, and the inputting the face image into the first feature extraction network to generate the first feature map can include the following steps:

[0040] The execution subject can input the face image into a plurality of serially connected second convolutional neural networks (CNNs) to output the first feature map. The number of the serially connected second convolutional neural networks is a preset number. This is not limited herein.

[0041] At step 204, a second feature map set representing local feature information of the face key regions is generated according to the first feature map and the probability map corresponding to each face key region.

[0042] In some embodiments, the execution subject can generate a second feature map set representing local feature information of the face key regions according to the first feature map and the probability map corresponding to each face key region. For example, the face key region is an eye key region, and the second feature map can represent eye feature information.

[0043] As an example, the execution subject can generate a second feature map set representing local feature information of the face key regions according to the first feature map and the probability map corresponding to each face key region, which can include the following steps:

[0044] First, for each probability map in the probability map corresponding to each face key region, a region determination step is performed:

[0045] First, a region in the probability map where the probability value is greater than or equal to a preset value is determined.

[0046] Second, for each region in the obtained region set, the following second feature map generation step is performed.

[0047] First, a feature map region in the first feature map corresponding to the region is determined.

[0048] Second, the feature map region is element-filled according to the size of the probability map to obtain a filled feature map as the second feature map.

[0049] At step 205, age information of the target organism is generated according to the second feature map set and the first feature map.

[0050] In some embodiments, the execution subject can generate age information of the target organism according to the second feature map set and the first feature map.

[0051] As an example, the execution subject can generate age information of the target organism according to the second feature map set and the first feature map, which can include the following steps:

[0052] In a first step, the execution subject can add the second feature maps in the second feature map set to obtain an added feature map. The added feature map can represent the overall feature information corresponding to the second feature map set.

[0053] In a second step, the execution subject merges the added feature map and the first feature map to obtain a merged feature map.

[0054] In a third step, the execution subject can input the merged feature map into a pre-trained convolutional neural network to output the age information of the target organism.

[0055] The age information generation method of some embodiments of the present disclosure can accurately generate the age information of the target organism. Specifically, the reason why the age information of the target organism cannot be accurately generated is that only the overall facial feature information of the face image to be recognized is used in the process of generating the age information of the target organism, resulting in inaccurate age information of the target organism. Therefore, the age information generation method of some embodiments of the present disclosure can first obtain a face image of a target organism to be age-identified, which is used to determine the age information of the target organism. Then, by inputting the face image into a face semantic segmentation network, a probability map corresponding to each facial key region can be accurately and efficiently generated. Then, the face image is input into a first feature extraction network to generate a first feature map representing the overall facial feature information. Furthermore, according to the first feature map and the probability map corresponding to each facial key region, a second feature map set representing the local feature information of each facial key region is accurately generated. Here, the generated second feature map set is used for subsequent generation of the age information of the target organism. Finally, based on multi-level feature information, not only the overall facial feature information of the face image is considered, but also the local facial feature information is considered, so that the subsequently generated age information is more accurate. Therefore, the age information generation method can accurately generate the age information of the target organism.

[0056] Further reference is made to Figure 3 FIG. 3 shows a flow 300 of another embodiment of the age information generation method according to the present disclosure. The age information generation method includes the following steps:

[0057] In step 301, a face image of a target organism to be age-identified is obtained.

[0058] In step 302, the face image is input into a pre-trained face semantic segmentation network to output a probability map corresponding to each facial key region in the face image.

[0059] Step 303, input the face image into the first feature extraction network to generate a first feature map.

[0060] In some embodiments, the specific implementation of steps 301-303 and the technical effects brought by them can refer to Figure 2 Steps 201-203 in the corresponding embodiments will not be repeated here.

[0061] Step 304, multiply each probability map corresponding to the face key area with the corresponding elements of the first feature map to obtain a set of multiplied feature maps as the second feature map set.

[0062] In some embodiments, the execution subject (for example Figure 1 The electronic device shown) can multiply each probability map corresponding to the face key area with the corresponding elements of the first feature map to obtain a set of multiplied feature maps as the second feature map set. Among them, the multiplied feature map includes the feature information of the corresponding face key area.

[0063] Step 305, superimpose each second feature map in the second feature map set to obtain a superimposed feature map.

[0064] In some embodiments, the execution subject can superimpose each second feature map in the second feature map set to obtain a superimposed feature map.

[0065] As an example, the feature map dimension of each second feature map in the second feature map set is: H*W*D. Among them, H*W is the resolution size of the second feature map. D is the number of channels of the second feature map. The number of each second feature map can be S. Then, the feature map dimension of the superimposed feature map is: H*W*(S*D).

[0066] Step 306, input the superimposed feature map into the pre-trained first convolutional layer to obtain a third feature map.

[0067] In some embodiments, the execution subject can input the superimposed feature map into the pre-trained first convolutional layer to obtain a third feature map.

[0068] Here, inputting the superimposed feature map into the convolutional layer can further extract the feature information of the superimposed feature map.

[0069] Step 307, add the third feature map to the first feature map to obtain an added feature map.

[0070] In some embodiments, the execution subject can add the third feature map and the first feature map corresponding to the feature map elements to obtain an added feature map. The added feature map includes feature information of the third feature map and the first feature map.

[0071] At step 308, the added feature map is input into the pre-trained second feature extraction network to output a fourth feature map.

[0072] In some embodiments, the execution subject can input the added feature map into the pre-trained second feature extraction network to output the fourth feature map.

[0073] Here, inputting the superimposed feature map into the second feature extraction network can further extract feature information of the added feature map.

[0074] At step 309, the fourth feature map is input into a plurality of serially connected fully connected layers to obtain the age information of the target organism.

[0075] In some embodiments, the execution subject can input the third feature map into a plurality of serially connected fully connected layers (FC) to obtain the age information of the target organism.

[0076] In some optional implementations of some embodiments, the convolutional network includes a second convolutional layer and an activation function layer connected in series. The activation function corresponding to the activation function layer can include but is not limited to one of the following: Sigmoid function (Sigmoid function), Rectified Linear Unit (ReLU).

[0077] It should be noted that all the networks and related feature map operations involved in the steps of generating age information from 301 to 309 constitute an overall age information generation network. For parameter training of the initial age information generation network, the following is specific:

[0078] The face semantic segmentation network in the initial age information generation network is a pre-trained network. During the parameter training of the initial age information generation network, the parameter values of the face semantic segmentation network are unchanged. The training sample set of the initial age information generation network can be pre-acquired. The training samples can include face images in RGB form and age labels corresponding to the face images.

[0079] Reference Figure 4 , Figure 4An example diagram of a network structure of the first feature extraction network is shown. The above-mentioned first feature extraction network includes three convolutional neural networks connected in series. That is, convolutional neural network 401, convolutional neural network 402, and convolutional neural network 403. Each convolutional neural network includes a convolutional layer and an activation function layer.

[0080] As can be seen from Figure 3 , compared with the description of some embodiments corresponding to Figure 2 , the flow 300 of the age information generation method in some embodiments corresponding to Figure 3 emphasizes more on the specific steps of how to generate age information according to the probability map corresponding to each facial key area and the first feature map. Thus, the scheme described in these embodiments reduces the amount of calculation for generating age information as much as possible through merging operation, convolution operation, addition operation, feature extraction operation and full connection operation on the basis of accurately generating age information, so as to facilitate application in reality.

[0081] Further referring to Figure 5 , as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of an age information generation device, which correspond to the method embodiments shown in Figure 2 , and the device can be specifically applied to various electronic devices.

[0082] As shown in Figure 5 , an age information generation device 500 includes an acquisition unit 501, a first input unit 502, a second input unit 503, a first generation unit 504, and a second generation unit 505. The acquisition unit 501 is configured to acquire a facial image of a target living body to be age-identified. The first input unit 502 is configured to input the facial image into a pre-trained facial semantic segmentation network to output a probability map corresponding to each facial key area in the facial image. The second input unit 503 is configured to input the facial image into a first feature extraction network to generate a first feature map. The first generation unit 504 is configured to generate a second feature map set representing local feature information of each facial key area according to the first feature map and the probability map corresponding to each facial key area. The second generation unit 505 is configured to generate age information of the target living body according to the second feature map set and the first feature map.

[0083] In some optional implementations of some embodiments, the first generation unit 504 in the above-mentioned device 500 can be further configured to: multiply the probability map corresponding to each facial key area with the first feature map respectively to obtain a multiplied feature map set as the second feature map set.

[0084] In some optional implementation of some embodiments, the second generating unit 505 in the apparatus 500 can be further configured to: perform feature map superposition on each second feature map in the second feature map set to obtain a superposed feature map; input the superposed feature map into the pre-trained first convolutional layer to obtain a third feature map; and add the third feature map and the first feature map to obtain an added feature map.

[0085] In some optional implementation of some embodiments, the second generating unit 505 in the apparatus 500 can be further configured to: input the added feature map into the pre-trained second feature extraction network to output a third feature map; and input the third feature map into a plurality of serially connected fully connected layers to obtain the age information of the target organism.

[0086] In some optional implementation of some embodiments, the first feature extraction network comprises a plurality of serially connected convolutional networks, and the second input unit 503 in the apparatus 500 can be further configured to further comprise: inputting the face image into the plurality of serially connected convolutional networks to output the first feature map.

[0087] In some optional implementation of some embodiments, the convolutional network comprises a second convolutional layer and an activation function layer connected in series.

[0088] In some optional implementation of some embodiments, the face image is a human face image, and each facial key region comprises an eye region, a mouth region, a nose region, and a forehead region.

[0089] It can be understood that the units described in the apparatus 500 correspond to the respective steps in the method described with reference to Figure 2 The operations, features, and advantages described above for the method also apply to the apparatus 500 and the units included therein, and thus are not described here again.

[0090] The following refers to Figure 6 which shows a structural schematic diagram of an electronic device (e.g., an electronic device in Figure 1 ) suitable for implementing some embodiments of the present disclosure. Figure 6 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of embodiments of the present disclosure.

[0091] As Figure 6As shown, the electronic device 600 can include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601 that can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 602 or loaded into a random access memory (RAM) 603 from a storage device 608. Various programs and data required for the operation of the electronic device 600 are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0092] Generally, the following devices can be connected to the I / O interface 605: input devices 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 609. The communication devices 609 can allow the electronic device 600 to communicate wirelessly or wired with other devices to exchange data. Although Figure 6 The electronic device 600 is shown with various devices, but it should be understood that all of the illustrated devices are not required, and more or fewer devices can alternatively be implemented. Figure 6 Each block shown in the flowcharts can represent a device, or a plurality of devices, as needed.

[0093] In particular, processes described above with reference to the flowcharts can be implemented as a computer software program according to some embodiments of the present disclosure. For example, some embodiments of the present disclosure include a computer program product including a computer program carried on a computer readable medium, the computer program containing program codes for performing the methods shown in the flowcharts. In some such embodiments, the computer program can be downloaded and installed from a network through the communication devices 609, or installed from the storage devices 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-described functions defined in the methods of some embodiments of the present disclosure are performed.

[0094] Note that the computer-readable medium or media used to provide embodiments of the present disclosure can be embodied by a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a computer-readable storage medium can include, but are not limited to, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the present disclosure, a computer-readable storage medium can be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, a computer-readable signal medium can include a computer-readable program code embodied in a computer-readable signal medium that can be transmitted from one place to another place for execution by, or to control the operation of, an instruction execution system, apparatus, or device. Also, a computer-readable medium can include any computer- readable medium for communicating, propagating, or transferring a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0095] In some embodiments, the client, server can communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication (e.g., communication networks) of any form or medium, such as the Internet or World Wide Web. Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), internetworks (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

[0096] The computer readable medium can be included in the electronic device; or can exist separately from the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquire a face image of a target organism to be age-identified; input the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image; input the face image into a first feature extraction network to generate a first feature map; generate a second feature map set representing local feature information of the face key regions according to the first feature map and the probability map corresponding to each face key region; and generate age information of the target organism according to the second feature map set and the first feature map.

[0097] Computer program code for carrying out operations of some embodiments of the disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0098] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of present disclosure. In this regard, each block in the flow diagrams and the block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0099] The units described in some embodiments of the present disclosure can be implemented in the form of software, or can be implemented in the form of hardware. The described units can also be arranged in a processor, for example, can be described as: a processor comprising an acquisition unit, a first input unit, a second input unit, a first generation unit and a second generation unit. Among them, the name of these units does not constitute a limitation to the unit itself in some cases, for example, the acquisition unit can also be described as "a unit for acquiring a face image of a target biological body to be age-identified".

[0100] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SOCs), complex programmable logic devices (CPLDs), etc.

[0101] The above description is merely some preferred embodiments of the present disclosure and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with each other to form a technical solution with similar functions to the technical features disclosed in the embodiments of the present disclosure (but not limited to).

Claims

1. An age information generation method, comprising: obtaining a face image of a target organism to be age-identified; inputting the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image, wherein an element in the probability map corresponding to a face key region with a probability greater than or equal to a predetermined value is an element corresponding to the face key region; inputting the face image into a first feature extraction network to generate a first feature map; generating a second feature map set representing local feature information of the face key regions according to the first feature map and the probability map corresponding to each face key region, comprising: determining a region in the probability map with a probability value greater than or equal to a preset value; determining a feature map region in the first feature map corresponding to the region; performing element padding on the feature map region according to the dimension size of the probability map to obtain a padded feature map as a second feature map; generating age information of the target organism according to the second feature map set and the first feature map, comprising: stacking each second feature map in the second feature map set to obtain a stacked feature map; inputting the stacked feature map into a pre-trained first convolutional layer to obtain a third feature map; adding the third feature map and the first feature map to obtain an added feature map, comprising: adding corresponding feature map elements of the third feature map and the first feature map to obtain the added feature map.

2. The method of claim 1, wherein, The generation of age information of the target organism according to the second feature map set and the first feature map further comprises: inputting the added feature map into a pre-trained second feature extraction network to output a fourth feature map; inputting the fourth feature map into a plurality of serially connected fully connected layers to obtain the age information of the target organism.

3. The method of claim 1, wherein, The first feature extraction network comprises a plurality of serially connected convolutional networks; and The inputting of the face image into the first feature extraction network to generate a first feature map comprises: inputting the face image into the plurality of serially connected convolutional networks to output the first feature map.

4. The method of claim 3, wherein, The convolutional network comprises a second convolutional layer and an activation function layer connected in series.

5. The method of claim 1, wherein, The face image is a human face image, and the face key regions include an eye region, a mouth region, a nose region, and a forehead region.

6. An age information generation apparatus, comprising: an obtaining unit configured to obtain a face image of a target organism to be age-identified; a first input unit configured to input the face image into a pre-trained face semantic segmentation network to output a probability map corresponding to each face key region in the face image, wherein an element in the probability map corresponding to a face key region with a probability greater than or equal to a predetermined value is an element corresponding to the face key region; a second input unit configured to input the face image into a first feature extraction network to generate a first feature map; The first generating unit is configured to generate a second feature map set representing local feature information of the face key regions according to the first feature map and the probability map corresponding to each face key region, including: determining a region in the probability map where a corresponding probability value is greater than or equal to a preset value; determining a feature map region in the first feature map corresponding to the region; performing element padding on the feature map region according to the size of the probability map to obtain a padded feature map as a second feature map; The second generating unit is configured to generate age information of the target biological body according to the second feature map set and the first feature map, including: superimposing each second feature map in the second feature map set to obtain a superimposed feature map; inputting the superimposed feature map into a pre-trained first convolutional layer to obtain a third feature map; adding the third feature map and the first feature map to obtain an added feature map, including: adding corresponding feature map elements of the third feature map and the first feature map to obtain the added feature map. 7.An electronic device, comprising: one or more processors; a memory device having stored thereon one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-5.

8. A computer readable medium having stored thereon a computer program, wherein, The program is executed by the processor to implement the method of any one of claims 1-5.

Images