CN113609966A - Method and device for generating training sample of face recognition system - Google Patents
Method and device for generating training sample of face recognition system Download PDFInfo
- Publication number
- CN113609966A CN113609966A CN202110886544.2A CN202110886544A CN113609966A CN 113609966 A CN113609966 A CN 113609966A CN 202110886544 A CN202110886544 A CN 202110886544A CN 113609966 A CN113609966 A CN 113609966A
- Authority
- CN
- China
- Prior art keywords
- image
- face
- preset
- target user
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides a generation method and a generation device of a training sample of a face recognition system, wherein the generation method is used for acquiring a face data image of a target user, and the face data image is a face image of a user wearing a mask; calling an image detector to detect the position information of the human face in the human face data image; cutting the face image according to the position information to obtain a target face image; adjusting the target face image to accord with a preset generation condition; and covering the confrontation sample mask image generated by the generator with a layer of the target face image which meets the preset generation condition to obtain the confrontation sample mask image of the target user. The embodiment of the invention can generate the confrontation sample mask image as the training sample of the face recognition system so as to improve the recognition precision of the face recognition system.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a method and a device for generating a training sample of a face recognition system.
Background
The application of face recognition systems in daily life is becoming more and more widespread. For example, face-swiping payment, office card swiping, face-swiping door access, and the like are performed by using the face recognition system.
If the target user wears accessories such as eyes, a hat, a sticker, etc. containing countermeasure information, it is possible for the recognition result of the face recognition system to recognize the target user as another user.
At present, common confrontation information is made into wearable accessories such as glasses, stickers and the like, and when a target user wears the accessory and appears in a camera, a face recognition system can identify the target user as other users by mistake. Particularly, with the occurrence of epidemic situations in recent years, people wearing the mask to go out become an essential part of daily life of people. The identification accuracy rate of the user wearing the mask also meets the requirement.
If the identification accuracy of a user wearing the mask by a face system is to be improved, training samples of the face identification system can be improved, but a method for generating a countersample mask is not available at present.
Disclosure of Invention
The embodiment of the invention aims to provide a method and a device for generating a training sample of a face recognition system, which are used for generating an confrontation sample mask image as the training sample of the face recognition system and improving the recognition accuracy of the face recognition system.
The invention discloses a method for generating a training sample of a face recognition system, which comprises the following steps:
acquiring a face data image of a target user, wherein the face data image is a face image of a wearing mask;
calling an image detector to detect the position information of the human face in the human face data image;
cutting the face image according to the position information to obtain a target face image;
adjusting the target face image to accord with a preset generation condition;
and covering the confrontation sample mask image generated by the generator with a layer of the target face image which meets the preset generation condition to obtain the confrontation sample mask image of the target user.
In some embodiments, further comprising:
training a preset face recognition model by using the confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user; and the similarity between the face data image of the interfering user and the confrontation sample mask image is smaller than a preset threshold value.
In some embodiments, the target user comprises a plurality of users, the method further comprising:
the face images of different users are classified into different folders, wherein the picture identifiers under each folder are the same.
In some embodiments, the preset generation condition is 224 × 224 pixels.
In some embodiments, the training of the preset face recognition model by using the confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interfering user comprises:
calculating a loss function according to the picture identification;
updating parameters of the generator with the loss function;
and training the preset face recognition model according to the updated confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user generated by the generator after the parameters are updated until the output value of the preset face model conforms to the specified range.
In some embodiments, the loss function enables the cosine similarity between the confrontation sample mask image of the target user and the face data image of the interfering user to meet a first preset value, and enables the cosine similarity between the confrontation sample mask image of the target user and the face data image of the target user to meet a second preset value.
In some embodiments, the generator is constructed using a convolutional neural network CNN.
In some embodiments, the step of covering the confrontation sample mask image generated by the generator with the target face image meeting the preset generation condition includes:
setting the target face image which accords with the preset generation condition in a preset canvas as a first image layer;
setting the confrontation sample mask image generated by the generator as a second image layer;
moving the second layer to overlap with the first layer;
setting the second layer as a top layer;
and setting the image parameters of the second image layer to be matched with the image parameters of the first image layer.
The invention also discloses a device for generating training samples of the face recognition system, which comprises:
the acquisition module is used for acquiring a face data image of a target user, wherein the face data image is a face image of a wearer wearing a mask;
the recognition module is used for calling an image detector to detect the position information of the human face in the human face data image;
the cutting module is used for cutting the face image according to the position information to obtain a target face image;
the adjusting module is used for adjusting the target face image to accord with a preset generating condition;
and the generation module is used for covering the confrontation sample mask image generated by the generator with the layer of the target face image which meets the preset generation condition to obtain the confrontation sample mask image of the target user.
In some embodiments, further comprising:
the model training module is used for training a preset face recognition model by utilizing an confrontation sample mask image of the target user, a face data image of the target user and a face data image of an interfering user; and the similarity between the face data image of the interfering user and the confrontation sample mask image is smaller than a preset threshold value.
The embodiment of the invention provides a method and a device for generating a training sample of a face recognition system, wherein the method comprises the steps of collecting a face data image of a target user, wherein the face data image is a face image of a user wearing a mask; calling an image detector to detect the position information of the human face in the human face data image; cutting the face image according to the position information to obtain a target face image; adjusting the target face image to accord with a preset generation condition; and covering the confrontation sample mask image generated by the generator with a layer of the target face image which meets the preset generation condition to obtain the confrontation sample mask image of the target user.
The embodiment of the invention has the following beneficial effects:
the confrontation sample mask image can be generated to be used as a training sample of the face recognition system, so that the recognition accuracy of the face recognition system is improved.
Of course, it is not necessary for any product or method of practicing the invention to achieve all of the above-described advantages at the same time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for generating a training sample of a face recognition system according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a device for generating training samples of a face recognition system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the field of face recognition, after a face image of a target user is recognized, a face recognition model can classify the face image into a correct ID so as to correctly recognize the target user. The face countermeasure sample is formed by adding a disturbance to a normal face, so that the face recognition model classifies the recognized face image into a wrong ID, a wrong user is recognized, and loss is caused to a target user. In order to avoid the false result recognized by the face recognition system, the recognition accuracy of the face recognition system needs to be improved.
The confrontation sample mask is an implementation mode of confrontation information, and in order to improve the accuracy of the recognition result of the face recognition system, a training sample is improved, so that the recognition model of the face recognition system is trained to improve the accuracy of recognition.
In order to overcome the defects in the prior art, the embodiment of the application discloses a method for generating a training sample of a face recognition system.
Referring to fig. 1, fig. 1 is a schematic flow chart of a method for generating a training sample of a face recognition system according to an embodiment of the present invention.
The invention discloses a method for generating a training sample of a face recognition system, which comprises the following steps:
s100, acquiring a face data image of a target user, wherein the face data image is a face image of a user wearing a mask;
the execution subject of the embodiment of the present invention may be a face recognition system, or a device, a processor, or a server that may execute each step in the embodiment of the present invention. The embodiment of the invention can be applied to computer vision related scenes, in particular to the improvement of training samples of recognition models used by deep learning technologies such as a face recognition system.
In this embodiment of the present invention, the target user includes a plurality of users, and the method further includes:
the face images of different users are classified into different folders, wherein the picture identifiers under each folder are the same.
In the embodiment of the invention, a face data set is collected: collecting a batch of face data, classifying the pictures with different IDs into different folders, wherein the IDs of the pictures in the same folder are the same.
In order to generate a training sample, a face image of a person wearing a mask is obtained. In the embodiment of the invention, the data images of the Reynna can be obtained in batches, can be directly obtained in a training set and a verification set of the face recognition system, and can also be obtained by collecting from a collecting device.
S200, calling an image detector to detect the position information of the face in the face data image;
s300, cutting the face image according to the position information to obtain a target face image;
s400, adjusting the target face image to accord with a preset generation condition;
in the embodiment of the invention, the face data image is preprocessed.
In the embodiment of the invention, the image detector can be used for carrying out face position recognition on the personal face data image to obtain position information, then the target face image is cut out, and then the target face image is adjusted to meet the preset generation condition, for example, the preset generation condition is 224x224 pixels. Of course, the preset generating conditions can be set according to actual needs, and can be set according to the requirements of the training samples.
And S500, covering the confrontation sample mask image generated by the generator with a layer of the target face image meeting the preset generation condition to obtain the confrontation sample mask image of the target user.
And after the target face image which meets the preset generation condition is obtained, the target face image is made into a confrontation sample mask image to be used as a training sample.
In the embodiment of the present invention, the step of covering the confrontation sample mask image generated by the generator with the layer of the target face image that meets the preset generation condition to obtain the confrontation sample mask image of the target user includes:
setting the target face image which accords with the preset generation condition in a preset canvas as a first image layer;
setting the confrontation sample mask image generated by the generator as a second image layer;
moving the second layer to overlap with the first layer;
setting the second layer as a top layer;
and setting the image parameters of the second image layer to be matched with the image parameters of the first image layer.
In the embodiment of the invention, the confrontation sample mask of the target user can be generated in a preset canvas of a preset tool such as drawing software.
The image parameters may include, among other things, sharpness, brightness, hue, and so on. The purpose is to make the image in the second layer and the image in the first layer look unobtrusive from a visual effect, and more likely to cause interference to a face recognition system when recognizing the images.
The size of the confrontation sample mask image generated by the generator also meets the preset generation condition, such as 224 × 224 pixels.
In practical use, a generator is constructed by using a Convolutional Neural Network (CNN), the generator inputs a 128-dimensional random vector and outputs an antagonistic sample mask image with the size of 224 multiplied by 224, and the antagonistic sample mask image and the real face data image are subjected to layer superposition to obtain an antagonistic sample mask image worn on an antagonistic sample mask.
In the embodiment of the invention, the larger the area of the pattern is, the greater the interference on a face recognition system is, the higher the value of the obtained training sample is, and the higher the recognition accuracy of the face recognition system after training is.
In the embodiment of the present invention, in order to make the confrontation sample mask image as the best training sample of the face recognition system, the method further includes:
training a preset face recognition model by using the confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user; and the similarity between the face data image of the interfering user and the confrontation sample mask image is smaller than a preset threshold value.
In some embodiments, the training of the preset face recognition model by using the confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interfering user comprises:
calculating a loss function according to the picture identification;
updating parameters of the generator with the loss function;
and training the preset face recognition model according to the updated confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user generated by the generator after the parameters are updated until the output value of the preset face model conforms to the specified range.
In some embodiments, the loss function enables the cosine similarity between the confrontation sample mask image of the target user and the face data image of the interfering user to meet a first preset value, and enables the cosine similarity between the confrontation sample mask image of the target user and the face data image of the target user to meet a second preset value.
In practical use, a trained face recognition model, such as arcface, is prepared, the input of the model is a picture with a fixed size of 224 × 224, and the output is a 256-dimensional vector. Selection of the loss function of the face recognition model: the similarity between the face image of the target user wearing the countersample mask and the face image of the interference user is higher, the similarity between the face image of the target user not wearing the mask is lower, and the cosine similarity can be selected as a loss function.
And inputting three types of face images, namely an confrontation sample mask image of a target user, a face data image of the target user and a face data image of an interfering user, into a face recognition model, and then calculating a loss function according to the ID label. And updating the parameters of the generator in the reverse direction by using a back propagation algorithm until the whole network converges. The parameters of the generator after network convergence are the optimal parameters of the confrontation sample mask, and the obtained confrontation sample mask image is the optimal training sample.
Therefore, the embodiment of the invention can generate the confrontation sample mask image as the training sample of the face recognition system so as to improve the recognition precision of the face recognition system.
Corresponding to the method embodiment, the embodiment of the invention also discloses a device for identifying the confrontation sample mask.
Referring to fig. 2, fig. 2 is a schematic structural diagram of a device for generating training samples of a face recognition system according to an embodiment of the present invention.
The invention discloses a device for generating training samples of a face recognition system, which comprises:
the system comprises an acquisition module 1, a display module and a display module, wherein the acquisition module is used for acquiring a face data image of a target user, and the face data image is a face image of a wearing mask;
the recognition module 2 is used for calling an image detector to detect the position information of the human face in the human face data image;
the cutting module 3 is used for cutting the face image according to the position information to obtain a target face image;
the adjusting module 4 is used for adjusting the target face image to accord with a preset generating condition;
and the generating module 5 is used for covering the confrontation sample mask image generated by the generator with the layer of the target face image which meets the preset generating condition to obtain the confrontation sample mask image of the target user.
In some embodiments, further comprising:
the model training module is used for training a preset face recognition model by utilizing an confrontation sample mask image of the target user, a face data image of the target user and a face data image of an interfering user; and the similarity between the face data image of the interfering user and the confrontation sample mask image is smaller than a preset threshold value.
In some embodiments, the target user comprises a plurality of users, and the obtaining means is further configured to:
the face images of different users are classified into different folders, wherein the picture identifiers under each folder are the same.
In some embodiments, the preset generation condition is 224 × 224 pixels.
In some embodiments, the training of the preset face recognition model by using the confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interfering user comprises:
calculating a loss function according to the picture identification;
updating parameters of the generator with the loss function;
and training the preset face recognition model according to the updated confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user generated by the generator after the parameters are updated until the output value of the preset face model conforms to the specified range.
In some embodiments, the loss function enables the cosine similarity between the confrontation sample mask image of the target user and the face data image of the interfering user to meet a first preset value, and enables the cosine similarity between the confrontation sample mask image of the target user and the face data image of the target user to meet a second preset value.
In some embodiments, the generator is constructed using a convolutional neural network CNN.
In some embodiments, the step of covering the confrontation sample mask image generated by the generator with the target face image meeting the preset generation condition includes:
setting the target face image which accords with the preset generation condition in a preset canvas as a first image layer;
setting the confrontation sample mask image generated by the generator as a second image layer;
moving the second layer to overlap with the first layer;
setting the second layer as a top layer;
and setting the image parameters of the second image layer to be matched with the image parameters of the first image layer.
Since the functions of each module correspond to each step in the foregoing method embodiment, the effect corresponding to the method can be achieved, and the functions of each module in the apparatus embodiment are not described herein again.
The generation device can generate the confrontation sample mask image as a training sample of the face recognition system so as to improve the recognition precision of the face recognition system.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a device includes one or more processors (CPUs), memory, and a bus. The device may also include input/output interfaces, network interfaces, and the like.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method for generating training samples of a face recognition system is characterized by comprising the following steps:
acquiring a face data image of a target user, wherein the face data image is a face image of a wearing mask;
calling an image detector to detect the position information of the human face in the human face data image;
cutting the face image according to the position information to obtain a target face image;
adjusting the target face image to accord with a preset generation condition;
and covering the confrontation sample mask image generated by the generator with a layer of the target face image which meets the preset generation condition to obtain the confrontation sample mask image of the target user.
2. The generation method according to claim 1, further comprising:
training a preset face recognition model by using the confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user; and the similarity between the face data image of the interfering user and the confrontation sample mask image is smaller than a preset threshold value.
3. The generation method of claim 1, wherein the target user comprises a plurality of users, the method further comprising:
the face images of different users are classified into different folders, wherein the picture identifiers under each folder are the same.
4. The generation method according to claim 1, characterized in that the preset generation condition is 224x224 pixels.
5. The generation method according to claim 2, wherein the training of the predetermined face recognition model using the confrontation sample mask image of the target user, the face data image of the target user, and the face data image of the interfering user comprises:
calculating a loss function according to the picture identification;
updating parameters of the generator with the loss function;
and training the preset face recognition model according to the updated confrontation sample mask image of the target user, the face data image of the target user and the face data image of the interference user generated by the generator after the parameters are updated until the output value of the preset face model conforms to the specified range.
6. The generation method according to claim 5, wherein the loss function is such that the cosine similarity between the confrontation sample mask image of the target user and the face data image of the interfering user meets a first preset value, and the cosine similarity between the confrontation sample mask image of the target user and the face data image of the target user meets a second preset value.
7. The generation method according to claim 1, characterized in that the generator is constructed using a Convolutional Neural Network (CNN).
8. The generation method according to claim 1, wherein the step of covering the confrontation sample mask image generated by the generator with the image layer of the target face image that meets the preset generation condition includes:
setting the target face image which accords with the preset generation condition in a preset canvas as a first image layer;
setting the confrontation sample mask image generated by the generator as a second image layer;
moving the second layer to overlap with the first layer;
setting the second layer as a top layer;
and setting the image parameters of the second image layer to be matched with the image parameters of the first image layer.
9. An apparatus for generating training samples for a face recognition system, comprising:
the acquisition module is used for acquiring a face data image of a target user, wherein the face data image is a face image of a wearer wearing a mask;
the recognition module is used for calling an image detector to detect the position information of the human face in the human face data image;
the cutting module is used for cutting the face image according to the position information to obtain a target face image;
the adjusting module is used for adjusting the target face image to accord with a preset generating condition;
and the generation module is used for covering the confrontation sample mask image generated by the generator with the layer of the target face image which meets the preset generation condition to obtain the confrontation sample mask image of the target user.
10. The generation apparatus according to claim 9, further comprising:
the model training module is used for training a preset face recognition model by utilizing an confrontation sample mask image of the target user, a face data image of the target user and a face data image of an interfering user; and the similarity between the face data image of the interfering user and the confrontation sample mask image is smaller than a preset threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110886544.2A CN113609966A (en) | 2021-08-03 | 2021-08-03 | Method and device for generating training sample of face recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110886544.2A CN113609966A (en) | 2021-08-03 | 2021-08-03 | Method and device for generating training sample of face recognition system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113609966A true CN113609966A (en) | 2021-11-05 |
Family
ID=78339314
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110886544.2A Pending CN113609966A (en) | 2021-08-03 | 2021-08-03 | Method and device for generating training sample of face recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113609966A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114898450A (en) * | 2022-07-14 | 2022-08-12 | 中国科学院自动化研究所 | Face confrontation mask sample generation method and system based on generation model |
-
2021
- 2021-08-03 CN CN202110886544.2A patent/CN113609966A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114898450A (en) * | 2022-07-14 | 2022-08-12 | 中国科学院自动化研究所 | Face confrontation mask sample generation method and system based on generation model |
| CN114898450B (en) * | 2022-07-14 | 2022-10-28 | 中国科学院自动化研究所 | Face confrontation mask sample generation method and system based on generation model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110147721B (en) | Three-dimensional face recognition method, model training method and device | |
| Pang et al. | Classifying discriminative features for blur detection | |
| CN108171207A (en) | Face identification method and device based on video sequence | |
| CN110428399B (en) | Method, apparatus, device and storage medium for detecting image | |
| CN111241989A (en) | Image recognition method and device and electronic equipment | |
| CN103902958A (en) | Method for face recognition | |
| CN111274947B (en) | Multi-task multi-thread face recognition method, system and storage medium | |
| CN110826610A (en) | Method and system for intelligently detecting whether dressed clothes of personnel are standard | |
| CN111488943A (en) | Face recognition method and device | |
| CN111597910A (en) | Face recognition method, face recognition device, terminal equipment and medium | |
| CN111931628B (en) | Training method and device of face recognition model and related equipment | |
| CN111881740A (en) | Face recognition method, face recognition device, electronic equipment and medium | |
| Hernandez-Ortega et al. | FaceQvec: Vector quality assessment for face biometrics based on ISO compliance | |
| CN113609966A (en) | Method and device for generating training sample of face recognition system | |
| CN110909685A (en) | Posture estimation method, device, equipment and storage medium | |
| CN113392763A (en) | Face recognition method, device and equipment | |
| WO2023124869A1 (en) | Liveness detection method, device and apparatus, and storage medium | |
| CN112200109A (en) | Face attribute recognition method, electronic device, and computer-readable storage medium | |
| CN115223022B (en) | Image processing method, device, storage medium and equipment | |
| CN111428670B (en) | Face detection method, face detection device, storage medium and equipment | |
| JP4510562B2 (en) | Circle center position detection method, apparatus, and program | |
| CN109858355B (en) | Image processing method and related product | |
| CN114511911A (en) | Face recognition method, device and equipment | |
| CN114627542A (en) | Eye movement position determination method and device, storage medium and electronic equipment | |
| CN112183284A (en) | Safety information verification and designated driving order receiving control method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211223 Address after: A111, 1f, building 3, No. 1, zone 1, Lize Zhongyuan, Wangjing emerging industrial zone, Chaoyang District, Beijing 100020 Applicant after: MIAOZHEN INFORMATION TECHNOLOGY Co.,Ltd. Address before: Floor 29, 30, 31, 32, No. 701, Yunjin Road, Xuhui District, Shanghai, 200030 Applicant before: Shanghai minglue artificial intelligence (Group) Co.,Ltd. |