CN112101296B - Face registration method, face verification method, device and system - Google Patents
Face registration method, face verification method, device and system Download PDFInfo
- Publication number
- CN112101296B CN112101296B CN202011095448.8A CN202011095448A CN112101296B CN 112101296 B CN112101296 B CN 112101296B CN 202011095448 A CN202011095448 A CN 202011095448A CN 112101296 B CN112101296 B CN 112101296B
- Authority
- CN
- China
- Prior art keywords
- face
- image
- registration image
- beautifying
- beauty
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 115
- 238000012795 verification Methods 0.000 title claims abstract description 75
- 238000012545 processing Methods 0.000 claims abstract description 100
- 238000001514 detection method Methods 0.000 claims abstract description 56
- 238000000605 extraction Methods 0.000 claims abstract description 27
- 230000003796 beauty Effects 0.000 claims description 55
- 230000008569 process Effects 0.000 claims description 53
- 238000012549 training Methods 0.000 claims description 35
- 230000003287 optical effect Effects 0.000 claims description 19
- 230000001815 facial effect Effects 0.000 description 14
- 238000004590 computer program Methods 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000002537 cosmetic Substances 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 102100022778 POC1 centriolar protein homolog A Human genes 0.000 description 1
- 101710125073 POC1 centriolar protein homolog A Proteins 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the application provides a face registration method, a face verification device and a face verification system, wherein in the face registration method, after a first face registration image is acquired, face beautifying detection is carried out on the first face registration image, whether a face in the first face registration image is subjected to face beautifying processing is judged, if the face in the first face registration image is subjected to face beautifying processing, face beautifying removal processing is carried out on the face, a second face registration image comprising the face subjected to face beautifying removal processing is acquired, face feature extraction is carried out on the first face registration image and the second face registration image respectively, and the extracted face features are stored in a face library. When the face registration is carried out on the first face registration image subjected to the face beautifying treatment, the face characteristics of the face beautifying and the face characteristics of the face beautifying removal are stored in the face library, and the face library can provide the face characteristics of the face beautifying removal, so that the accuracy of face verification is improved.
Description
Technical Field
The present disclosure relates to the field of intelligent detection technologies, and in particular, to a face registration method, a face verification method, a device, and a system.
Background
With the continuous development of intelligent detection technology, people's life, work are more and more separated from electronic equipment (such as cell-phone, panel computer, entrance guard's equipment etc.), especially application of face verification technology in electronic equipment, for example, face unblock, face payment, entrance guard's equipment is automatic to be released etc. has brought very big facility for people's life and work.
At present, the face verification method mainly comprises the following steps: the face image to be verified is obtained, face feature extraction is carried out on the face image to be verified, the face features extracted are matched with the face features stored in the face library, if the matched face features exist in the face library, the face image to be verified is confirmed to pass verification, and further, the operations of unlocking, payment, access control equipment release and the like can be carried out. The face database is obtained by face registration in advance, the face registration refers to the process of storing face features in face registration images into the face database, the face registration images refer to personal credentials, life photos, headshot and the like of users, and the face registration images are often provided by the users.
However, the face registration image standards provided by the users cannot be unified, and some face registration images provided by the users are subjected to face beautifying treatment, and the face characteristics after face beautifying treatment have a larger difference from the actual face characteristics, so that the accuracy of face verification is lower.
Disclosure of Invention
The embodiment of the application aims to provide a face registration method, a face verification device and a face verification system so as to improve the accuracy of face verification. The specific technical scheme is as follows:
in a first aspect, an embodiment of the present application provides a face registration method, where the method includes:
acquiring a first face registration image;
face beautifying detection is carried out on the first face registration image, and whether the face in the first face registration image is subjected to face beautifying treatment is judged;
if yes, carrying out face beautifying removal processing on the face to obtain a second face registration image of the face after the face beautifying removal processing;
and respectively extracting the face features of the first face registration image and the second face registration image, and storing the extracted face features into a face library.
Optionally, the step of performing face beautifying detection on the first face registration image to determine whether the face in the first face registration image is subjected to face beautifying processing includes:
and inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment.
Optionally, the step of performing a face-beautifying removal process on the face to obtain a second face registration image including the face after the face-beautifying removal process includes:
inputting the first face registration image into a face beautifying removal model trained in advance, and carrying out face beautifying removal processing on the face in the first face registration image by utilizing the face beautifying removal model to obtain a second face registration image comprising the face after the face beautifying removal processing, wherein the face beautifying removal model is a deep network model obtained by training by taking the face image subjected to face beautifying processing as a training sample and taking the face image not subjected to face beautifying processing as a training target.
In a second aspect, an embodiment of the present application provides a face verification method, where the method includes:
acquiring a face image to be verified;
extracting facial features of the facial image to be verified;
matching the extracted face features with each face feature stored in a face library, wherein the face library at least stores face features obtained by extracting the face features of a first face registration image, and if face beautifying detection is performed on the first face registration image to determine that the first face registration image is subjected to face beautifying processing, the face library also stores face features obtained by extracting the face features of a second face registration image, and the second face registration image is obtained by performing face beautifying removal processing on the first face registration image;
If the face features matched with the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes the verification.
Optionally, the step of acquiring the face image to be verified includes:
receiving a face image to be verified, which is acquired by a face acquisition unit on access control equipment;
after the step of determining that the face image to be verified passes verification if the face features matched with the extracted face features exist in the face library, the method further comprises:
and sending a release notice to the access control equipment to inform the access control equipment to release.
Optionally, the face beautifying detection is performed on the first face registration image in the following manner:
and inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment.
Optionally, the face-beautifying removal process is performed on the first face image in the following manner, so as to obtain a second face registration image:
inputting the first face registration image into a face beautifying removal model trained in advance, and carrying out face beautifying removal processing on the face in the first face registration image by utilizing the face beautifying removal model to obtain a second face registration image comprising the face after the face beautifying removal processing, wherein the face beautifying removal model is a deep network model obtained by training by taking the face image subjected to face beautifying processing as a training sample and taking the face image not subjected to face beautifying processing as a training target.
In a third aspect, an embodiment of the present application provides a face registration apparatus, including:
the acquisition module is used for acquiring a first face registration image;
the detection module is used for carrying out face beautifying detection on the first face registration image and judging whether the face in the first face registration image is subjected to face beautifying treatment or not;
the beauty Yan Quchu module is used for performing beauty removal processing on the face if the detection result of the detection module is that the face in the first face registration image is subjected to the beauty processing, so as to obtain a second face registration image of the face after the beauty removal processing;
the feature extraction module is used for extracting the face features of the first face registration image and the second face registration image respectively and storing the extracted face features into the face library.
In a fourth aspect, an embodiment of the present application provides a face verification apparatus, including:
the acquisition module is used for acquiring the face image to be verified;
the feature extraction module is used for extracting the face features of the face image to be verified;
the matching module is used for matching the extracted face features with each face feature stored in the face database, wherein the face database is obtained based on the face registration method provided by the first aspect;
And the verification module is used for determining that the face image to be verified passes verification if the face features matched with the extracted face features exist in the face library.
In a fifth aspect, embodiments of the present application provide an electronic device comprising a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: the method provided in the first aspect of the present application is implemented.
In a sixth aspect, embodiments of the present application provide a machine-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, implement the method provided by the first aspect of embodiments of the present application.
In a seventh aspect, embodiments of the present application also provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform a method embodying the first aspect of the embodiments of the present application.
In an eighth aspect, embodiments of the present application provide an electronic device comprising a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: the method provided in the second aspect of the present application is implemented.
In a ninth aspect, embodiments of the present application provide a machine-readable storage medium storing machine-executable instructions that, when invoked and executed by a processor, implement the method provided by the second aspect of embodiments of the present application.
In a tenth aspect, embodiments of the present application also provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform a method embodying the second aspect provided by embodiments of the present application.
In an eleventh aspect, an embodiment of the present application provides a face verification system, including a face collection device and an electronic device;
the face acquisition equipment is used for acquiring a face image to be verified and sending the face image to be verified to the electronic equipment;
the electronic equipment is used for receiving the face image to be verified sent by the face acquisition equipment and extracting face characteristics of the face image to be verified; matching the extracted face features with each face feature stored in a face library, wherein the face library at least stores face features obtained by extracting the face features of a first face registration image, and if face beautifying detection is performed on the first face registration image to determine that the first face registration image is subjected to face beautifying processing, the face library also stores face features obtained by extracting the face features of a second face registration image, and the second face registration image is obtained by performing face beautifying removal processing on the first face registration image; if the face features matched with the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes the verification.
The embodiment of the application provides a face registration method, a face verification device and a face verification system, wherein in the face registration method, after a first face registration image is acquired, face beautifying detection is carried out on the first face registration image, whether a face in the first face registration image is subjected to face beautifying processing is judged, if the face in the first face registration image is subjected to face beautifying processing, face beautifying removal processing is carried out on the face, a second face registration image comprising the face subjected to face beautifying removal processing is acquired, face feature extraction is carried out on the first face registration image and the second face registration image respectively, and the extracted face features are stored in a face library.
Therefore, when the face registration image subjected to the face treatment is subjected to the face registration, the face characteristics of the face subjected to the face treatment are stored in the face library, and the face characteristics of the face removed are stored, namely, the face characteristics in the face library are expanded, and the face characteristics of the face removal are provided, so that the accuracy of face verification is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other embodiments may be obtained according to these drawings without inventive effort to a person skilled in the art.
Fig. 1 is a schematic flow chart of a face registration method according to an embodiment of the present application;
fig. 2 is another flow chart of a face registration method according to an embodiment of the present application;
fig. 3 is a schematic flow chart of a face verification method according to an embodiment of the present application;
fig. 4 is another flow chart of a face verification method according to an embodiment of the present application;
fig. 5 is a flowchart of a method for face registration according to an embodiment of the present application;
fig. 6 is a flowchart of a method for face verification according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a face registration device according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of a face verification device according to an embodiment of the present application;
FIG. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure;
Fig. 10 is a schematic structural diagram of another electronic device according to an embodiment of the present application;
fig. 11 is a schematic structural diagram of a face verification system according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
In order to improve accuracy of face verification, the embodiment of the application provides a face registration method, a face verification device and a face verification system. Next, first, a face registration method and a face verification method provided in the embodiments of the present application are described. The face registration method and the face verification method provided by the embodiment of the application can be applied to intelligent equipment such as mobile phones, tablet computers and access control equipment, and can also be applied to background servers, are not particularly limited, and are hereinafter collectively called as electronic equipment. The face registration method and the face verification method provided by the embodiment of the application can be realized by at least one of software, a hardware circuit and a logic circuit arranged in the electronic equipment.
As shown in fig. 1, a procedure of the face registration method provided in the embodiment of the present application may include the following steps.
S101, acquiring a first face registration image.
S102, face beautifying detection is carried out on the first face registration image, and whether the face in the first face registration image is subjected to face beautifying processing is judged.
And S103, if the face in the first face registration image is subjected to the face beautifying treatment, carrying out face beautifying removal treatment on the face to obtain a second face registration image comprising the face subjected to the face beautifying removal treatment.
S104, face feature extraction is carried out on the first face registration image and the second face registration image respectively, and the extracted face features are stored in a face library.
By applying the embodiment of the application, if the first face registration image is an image after the face beautifying process, the face beautifying removing process is performed on the face in the image to obtain the second face registration image, that is, the face in the first face registration image is the face after the face beautifying process, and the face in the second face registration image is the face after the face beautifying removing process, the extracted face features include the face features of the face beautifying process and the face features of the face beautifying removing process, so that when the face registration is performed on the face registration image of the first face subjected to the face beautifying process, the face features of the face beautifying process are stored in the face library, and the face features of the face beautifying removing process are also stored, namely the face features in the face library are expanded, and the face features of the face beautifying removing process are provided, so that the accuracy of face verification is improved.
The first face registration image refers to personal credentials (such as identity card, work license), living photo, big head photo and the like provided by a user when the user performs face registration, because the first face registration image is autonomously provided by the user, the format and standard of the first face registration images provided by different users cannot be unified, some of the first face registration images provided by the user are not subjected to face beautifying treatment, some of the first face registration images provided by the user are subjected to face beautifying treatment, and the face beautifying treatment refers to that the user adjusts the facial features, facial contours and the like of the face by using image repairing software, and the first face registration images subjected to face beautifying treatment easily change the identity of the user and influence the accuracy of face verification.
In the embodiment of the application, after the first face registration image is acquired, face beautifying detection is performed on the first face registration image, and whether the face in the first face registration image is subjected to face beautifying processing is judged. Since the face beautifying process is to adjust the facial feature shape, facial contour, etc. of the face, these adjustments will change the optical flow field at the corresponding position in the image, so the face beautifying process can be performed on the first face registration image by means of optical flow field prediction, if some positions in the first face registration image have obvious optical flow field changes, it is indicated that these positions have undergone operations such as image stretching/warping, that is, it can be determined that the face in the first face registration image has undergone the face beautifying process.
Alternatively, S102 may specifically be: and inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment.
In one implementation manner of the embodiment of the present application, the optical flow field prediction may be performed by using a face liquefaction detector (FALdetector), which is a detection tool for face liquefaction and beauty treatment, and can detect the deformation of the facial features and the adjustment of the facial contours in the face image. Specifically, the face beautifying detection method by using the face liquefaction detector is as follows: the first face registration image is input into a face liquefaction detector, and the face liquefaction detector can predict an optical flow field of the first face registration image and indicate the stretched/distorted area and degree of the image, so that whether the face in the first face registration image is subjected to face beautifying treatment can be judged. Of course, in a specific implementation, besides the fact that the face liquefaction detector is used for carrying out optical flow field prediction to determine whether the face in the first face registration image is subjected to the beautifying process or not, other optical flow field prediction methods can be adopted to carry out optical flow field prediction on the first face registration image, and whether the face in the first face registration image is subjected to the beautifying process or not is determined according to an optical flow field prediction result.
In another implementation manner of the embodiment of the present application, a two-class image type recognition model may be obtained in advance by training based on a large number of face sample images subjected to the face treatment and a large number of face sample images not subjected to the face treatment, where the model is a deep neural network model, and the output is that the images are subjected to the face treatment and the images are not subjected to the face treatment. In this way, the first face registration image is input to the model, and the model can directly output the result of recognition of whether the first face registration image is a face-beautifying processed image or a face-non-face-beautifying processed image. The training process of the image type recognition model is the same as that of the traditional two-classification deep neural network model, and is not repeated here.
In addition to the above-mentioned face beautifying detection mode, whether the face is subjected to beautifying treatment can also be detected through face feature recognition, comparison of face feature recognition and normal face features and other modes, and whether the face is subjected to beautifying treatment can be detected in a mode which belongs to the protection scope of the application and is not described in detail herein.
If the face in the first face registration image is detected to be subjected to face beautifying treatment, face beautifying removal treatment is required to be carried out on the face in order to improve the accuracy of face verification, a second face registration image comprising the face subjected to face beautifying removal treatment is obtained, and the face subjected to face beautifying removal treatment is closer to a normal face.
Optionally, S103 may specifically be: inputting the first face registration image into a face beautifying removal model trained in advance, and carrying out face beautifying removal processing on the face in the first face registration image by utilizing the face beautifying removal model to obtain a second face registration image comprising the face after the face beautifying removal processing, wherein the face beautifying removal model is a deep network model obtained by training by taking the face image subjected to face beautifying processing as a training sample and taking the face image not subjected to face beautifying processing as a training target.
In an implementation manner of the embodiment of the present application, a face-beautifying removal model that is trained by using a face image that is subjected to face-beautifying treatment as a training sample and a face image that is not subjected to face-beautifying treatment as a training target may be used to perform face-beautifying removal processing on a first face registration image, where the model may use a neural network framework such as Pix2 Pix.
Specifically, the training process of the face beautifying removal model comprises the following steps: inputting the face image subjected to the face beautifying treatment into an initial network model, outputting a face image by the network model, comparing the face image output by the network model with a face image not subjected to the face beautifying treatment to obtain difference information, and adjusting parameters of the initial network model based on the difference information until the iteration times of the initial network model reach preset times or the difference information is smaller than a preset threshold value, and stopping training to obtain the face beautifying removal model. The face image which is not subjected to the face beautifying treatment and the face image which is subjected to the face beautifying treatment have a corresponding relation, and generally, the face image which is not subjected to the face beautifying treatment can be obtained, and then the face image is subjected to the face beautifying treatment (comprising facial feature shape modification, facial outline adjustment and the like) through a picture trimming software, so that the corresponding face image which is subjected to the face beautifying treatment is obtained.
In another implementation manner of the embodiment of the present application, since the position and the degree of face beautification are detected during face beautification detection, according to the position and the degree of face beautification, the eyes can be subjected to beautification removal processing in a manner of reducing by one time according to the corresponding inverse processing manner, for example, the eyes can be amplified by one time in the beautification processing, and further the processing of face beautification Yan Quchu can be realized.
Besides the above-mentioned ways of performing the face-beautifying removing process, other ways of performing the face-beautifying removing process all belong to the protection scope of the present application, and are not described here in detail.
The face after the face removing treatment may be a plain face or a face after makeup, which is not limited herein, and is specifically related to the face removing policy selected for the face removing treatment.
After the second face registration image is obtained, face feature extraction is carried out on the first face registration image and the second face registration image respectively, the extracted face features are stored in a face library, namely, when the face registration is carried out on the first face registration image subjected to face beautifying treatment, the face features of the face beautifying and face feature removal are stored in the face library, namely, the face features in the face library are expanded, and the face features of the face removal are provided, so that the accuracy of face verification is improved.
The face feature extraction method can adopt a feature extraction method based on a deep neural network, and the first face registration image and the second face registration image are respectively input into the deep neural network, so that the respective face features can be extracted. The feature extraction method based on pixels can also be adopted, the face areas in the first face registration image and the second face registration image are identified based on the pixels, and the pixel features of the face areas are extracted as the face features. Of course, other ways of implementing face feature extraction also belong to the protection scope of the embodiments of the present application, and are not described here in detail.
Based on the embodiment shown in fig. 1, another flow of the face registration method provided in the embodiment of the present application, as shown in fig. 2, may include the following steps.
S201, a first face registration image is acquired.
S202, face beautifying detection is carried out on the first face registration image, whether the face in the first face registration image is subjected to face beautifying processing is judged, if yes, S203 to S204 are executed, and otherwise S205 is executed.
And S203, performing face beautifying removal processing on the face to obtain a second face registration image comprising the face after the face beautifying removal processing.
S204, face feature extraction is carried out on the first face registration image and the second face registration image respectively, and the extracted face features are stored in a face library.
S205, extracting the face features of the first face registration image, and storing the extracted face features into a face library.
In an implementation manner of the embodiment of the present application, if the face in the first face registration image is not subjected to the face beautifying process, it is indicated that the face in the first face registration image is a normal face, face feature extraction may be directly performed on the first face registration image, and the extracted face feature is stored in the face library.
In another implementation manner of the embodiment of the present application, in order to cope with more complicated situations, a face library is extended, if a face in a first face registration image is not subjected to face beautifying processing, special processing may be performed on the first face registration image, for example, cosmetic processing, face beautifying processing, etc. are performed on a face in the first face registration image, so as to obtain a new face registration image, face feature extraction is performed on the first face registration image and the new face registration image, and the extracted face feature is stored in the face library. The treatment modes of the cosmetic treatment, the beauty treatment and the like can be traditional treatment modes, and are not repeated here.
As shown in fig. 3, a flow of the face verification method provided in the embodiment of the present application may include the following steps.
S301, acquiring a face image to be verified.
S302, face feature extraction is carried out on the face image to be verified.
S303, matching the extracted face features with each face feature stored in a face library.
The face library at least stores face features obtained by extracting face features of the first face registration image, and if face beautifying detection is performed on the first face registration image to determine that the first face registration image is subjected to face beautifying processing, the face library also stores face features obtained by extracting face features of the second face registration image, wherein the second face registration image is obtained by performing face beautifying removal processing on the first face registration image.
The generation manner of the face database may be referred to the foregoing description of fig. 1 and fig. 2, and will not be described herein.
When a plurality of face features are stored in the face library, the extracted face features may be sequentially matched with each face feature stored in the face library, or the extracted face features may be respectively matched with each face feature stored in the face library in parallel. And if the extracted face features are matched with each face feature stored in the face library in sequence, and the face features matched with the face features obtained by extracting the face features of the face image to be verified are determined to exist in the matching process, the matching can be continued or can be terminated.
S304, if the face characteristics matched with the face characteristics obtained by extracting the face characteristics of the face image to be verified exist in the face library, determining that the face image to be verified passes the verification.
By applying the embodiment of the application, when the face registration is carried out on the first face registration image subjected to face beautifying treatment, the face characteristics of the face can be stored in the face library, and the face characteristics of the face removed are stored, namely, the face characteristics of the face in the face library are expanded, the face characteristics of the face removed are provided, that is, the face characteristics of the face in the face library are more perfect, and the face characteristics of the face removed which are more beneficial to face verification are included, so that the face can be more accurately identified when the face verification is carried out, and the accuracy of the face verification is improved.
The face image to be verified is a face image acquired in real time, and can be a face image shot in real time when a face appears in a shooting area by a face acquisition unit on a mobile phone, a tablet personal computer or access control equipment.
And after the face image to be verified is obtained, extracting the face characteristics of the face image to be verified. The specific face feature extraction mode can adopt a feature extraction mode based on a deep neural network, and the face image to be verified is input into the deep neural network, so that the face features can be extracted; the feature extraction method based on the pixels can also be adopted, the face region in the face image to be verified is identified based on the pixels, and the pixel features of the face region are extracted as the face features. Of course, other ways of implementing face feature extraction also belong to the protection scope of the embodiments of the present application, and are not described here in detail.
After the face features of the face image to be verified are extracted, the extracted face features are matched with each face feature stored in a face library, namely the extracted face features are compared with the face features stored in the face library, the face features stored in the face library are obtained by face registration according to the method of the embodiment shown in fig. 1 or fig. 2, if the similarity reaches a certain degree (more than a certain threshold), the face features are considered to be successfully matched, and once the face features successfully matched (whether the face features are face features subjected to face beautifying treatment or face features not subjected to face beautifying treatment) in the face library, the face image to be verified can be determined to pass verification.
Based on the embodiment shown in fig. 3, another flow of the face verification method provided in the embodiment of the present application, as shown in fig. 4, may include the following steps.
S401, receiving a face image to be verified, which is acquired by a face acquisition unit on the access control equipment.
S402, face feature extraction is carried out on the face image to be verified.
S403, matching the extracted face features with each face feature stored in a face database.
The face library at least stores face features obtained by extracting face features of the first face registration image, and if face beautifying detection is performed on the first face registration image to determine that the first face registration image is subjected to face beautifying processing, the face library also stores face features obtained by extracting face features of the second face registration image, wherein the second face registration image is obtained by performing face beautifying removal processing on the first face registration image.
The generation manner of the face database may be referred to the foregoing description of fig. 1 and fig. 2, and will not be described herein. .
S404, if the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes the verification.
S405, sending a release notice to the access control equipment to inform the access control equipment of release.
In a specific application of the embodiment of the application, the application scene is a face verification scene of an access control device in a community/park, the acquired face image to be verified is collected and sent by a face collection unit on the access control device, and accordingly, after the fact that the face image to be verified passes verification is confirmed, a release notification can be sent to the access control device, the release notification can be specifically a data message, after the access control device receives the release notification, the access control device knows that release operation is to be carried out, and a series of release processing such as rod lifting and door opening can be controlled.
In summary, in the application of face verification provided in the embodiment of the present application, the application mainly includes two parts, namely face registration and face verification. The face registration process is shown in fig. 5, and includes the following steps:
s501, obtaining a first certificate uploaded by a user.
The operation usually requires the user to provide the certificate of the electronic version or the user to operate independently on the application program, and most of the certificates provided or uploaded by the user are subjected to the beauty treatment by various graphic repair software, mainly relating to the adjustment of five sense organs and facial forms; the uploaded credential is marked as a first credential.
S502, performing beauty detection on the first certificate photo.
The detection that the face liquefaction was handled is handled to the face is main in this embodiment, including facial feature deformation and facial contour adjustment, adopts face liquefaction detector, inputs first certificate of face liquefaction detector, and face liquefaction detector can predict the light flow field, instructs first certificate to be stretched/crooked region and degree to whether judge the face and handle through the face.
S503, if the beauty detection judges that the beauty treatment exists, the beauty removal treatment is carried out, and the treated certificate is marked as a second certificate.
In the embodiment of the application, the beauty removal is mainly realized based on the generation of the countermeasure network, a Pix2Pix frame is to be adopted, and a face liquefying tool of the mapping software is used for generating a face image for the beauty treatment, and the facial feature shape modification (including eyes, nose and mouth) and the adjustment of the face outline are involved. And constructing pairing training data by using the generated face image subjected to the beautifying treatment and the original face image not subjected to the beautifying treatment, and then training a network model. And carrying out liquefaction reduction treatment on the input first certificate by using the model to realize face beautifying removal.
S504, face registration is carried out on the first certificate photo and the second certificate photo respectively.
The face features of the first certificate photo and the second certificate photo are extracted and stored in a face library respectively, and the two face features can be marked as different registered versions of the same identity in the face library.
The face verification process is shown in fig. 6, and includes the following steps:
s601, a face acquisition unit on the access control equipment acquires face images.
The face image processing method and device are mainly applied to access control equipment manufacturers, the face acquisition unit is a camera on access control equipment, and the acquired face image does not have any beautifying processing.
S602, face feature extraction.
In the embodiment of the application, the face feature extraction can be performed on the acquired face image by using a face recognition model.
S603, face comparison.
The comparison process is to compare the extracted face features with the face features of the first certificate photo and the face features of the second certificate photo of the same person respectively, and if only one version of the face features and the extracted face features reach the requirement (exceeding the preset threshold), the face verification is considered to pass.
Corresponding to the above method embodiment, the embodiment of the present application provides a face registration device, as shown in fig. 7, which may include:
An acquiring module 710, configured to acquire a first face registration image;
the detection module 720 is configured to perform face beautifying detection on the first face registration image, and determine whether a face in the first face registration image is subjected to face beautifying processing;
the face-beautifying Yan Quchu module 730 is configured to perform face-beautifying removal processing on the face if the detection result of the detection module 720 is that the face in the first face-beautifying registered image is subjected to face-beautifying processing, so as to obtain a second face-beautifying registered image including the face after the face-beautifying removal processing;
the feature extraction module 740 is configured to extract face features of the first face registration image and the second face registration image, and store the extracted face features in the face database.
Optionally, the detection module 720 may specifically be configured to:
and inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment.
Optionally, the module 730 of america Yan Quchu may be specifically configured to:
inputting the first face registration image into a face beautifying removal model trained in advance, and carrying out face beautifying removal processing on the face in the first face registration image by utilizing the face beautifying removal model to obtain a second face registration image comprising the face after the face beautifying removal processing, wherein the face beautifying removal model is a deep network model obtained by training by taking the face image subjected to face beautifying processing as a training sample and taking the face image not subjected to face beautifying processing as a training target.
By applying the embodiment of the application, if the first face registration image is an image after the face beautifying process, the face beautifying removing process is performed on the face in the image to obtain the second face registration image, that is, the face in the first face registration image is the face after the face beautifying process, and the face in the second face registration image is the face after the face beautifying removing process, the extracted face features include the face features of the face beautifying process and the face features of the face beautifying removing process, so that when the face registration is performed on the face registration image of the first face subjected to the face beautifying process, the face features of the face beautifying process are stored in the face library, and the face features of the face beautifying removing process are also stored, namely the face features in the face library are expanded, and the face features of the face beautifying removing process are provided, so that the accuracy of face verification is improved.
The embodiment of the application also provides a face verification device, as shown in fig. 8, the device may include:
an acquiring module 810, configured to acquire a face image to be verified;
the feature extraction module 820 is configured to perform face feature extraction on the face image to be verified;
the matching module 830 is configured to match the extracted face feature with each face feature stored in the face database, where the face database stores at least a face feature obtained by extracting a face feature of the first face registration image, and if the first face registration image is determined to be subjected to face beautifying through face beautifying detection, the face database also stores a face feature obtained by extracting a face feature of the second face registration image, where the second face registration image is obtained by performing face beautifying removal processing on the first face registration image;
And the verification module 840 is configured to determine that the face image to be verified passes verification if there are face features in the face library that match with face features obtained by extracting the face features of the face image to be verified.
Optionally, the obtaining module 810 may specifically be configured to: receiving a face image to be verified, which is acquired by a face acquisition unit on access control equipment;
the apparatus may further include:
and the sending module is used for sending a release notice to the access control equipment so as to inform the access control equipment of releasing.
By applying the embodiment of the application, when the face registration is carried out on the first face registration image subjected to face beautifying treatment, the face characteristics of the face can be stored in the face library, and the face characteristics of the face removed are stored, namely, the face characteristics of the face in the face library are expanded, the face characteristics of the face removed are provided, that is, the face characteristics of the face in the face library are more perfect, and the face characteristics of the face removed which are more beneficial to face verification are included, so that the face can be more accurately identified when the face verification is carried out, and the accuracy of the face verification is improved.
The present embodiments also provide an electronic device, as shown in fig. 9, including a processor 901 and a machine-readable storage medium 902, the machine-readable storage medium 902 storing machine-executable instructions capable of being executed by the processor 901, the processor 901 being caused by the machine-executable instructions to: the face registration method is realized.
The present embodiments also provide an electronic device, as shown in fig. 10, including a processor 1001 and a machine-readable storage medium 1002, the machine-readable storage medium 1002 storing machine-executable instructions capable of being executed by the processor 1001, the processor 1001 being caused by the machine-executable instructions to: the face verification method is achieved.
The machine-readable storage medium may include RAM (Random Access Memory ) or NVM (Non-Volatile Memory), such as at least one magnetic disk Memory. In the alternative, the machine-readable storage medium may also be at least one memory device located remotely from the processor.
The processor may be a general-purpose processor, including a CPU (Central Processing Unit ), NP (Network Processor, network processor), etc.; but also DSP (Digital Signal Processing, digital signal processor), ASIC (Application Specific Integrated Circuit ), FPGA (Field-Programmable Gate Array, field programmable gate array) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
Data transfer between the machine-readable storage medium 902 and the processor 901, between the machine-readable storage medium 1002 and the processor 1001 may be by way of wired or wireless connections, and the electronic device may communicate with other devices via wired or wireless communication interfaces. The examples shown in fig. 9 and 10 are merely examples of data transmission between the processor and the machine-readable storage medium via the bus, and are not intended to be limiting of the specific connection.
The embodiment of the application also provides a machine-readable storage medium which stores machine-executable instructions and realizes the face registration method when being called and executed by a processor.
The embodiments of the present application also provide a computer program product containing instructions that, when run on a computer, cause the computer to perform implementing the face registration method described above.
The embodiment of the application also provides a machine-readable storage medium which stores machine-executable instructions and realizes the face verification method when being called and executed by a processor.
The embodiments of the present application also provide a computer program product comprising instructions that, when run on a computer, cause the computer to perform implementing the above-described face verification method.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, DSL (Digital Subscriber Line, digital subscriber line)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy Disk, a hard Disk, a magnetic tape), an optical medium (e.g., a DVD (Digital Versatile Disc, digital versatile Disk)), or a semiconductor medium (e.g., an SSD (Solid State Disk)), or the like.
The embodiment of the application also provides a face verification system, as shown in fig. 11, which comprises a face acquisition device 1101 and an electronic device 1102;
the face acquisition device 1101 is configured to acquire a face image to be verified, and send the face image to be verified to the electronic device 1102;
the electronic device 1102 is configured to receive a face image to be verified sent by the face acquisition device 1101, and extract face features of the face image to be verified; matching the extracted face features with each face feature stored in a face library, wherein the face library at least stores face features obtained by extracting the face features of a first face registration image, and if face beautifying detection is performed on the first face registration image to determine that the first face registration image is subjected to face beautifying processing, the face library also stores face features obtained by extracting the face features of a second face registration image, and the second face registration image is obtained by performing face beautifying removal processing on the first face registration image; if the face features matched with the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes the verification.
The face acquisition equipment can be a mobile phone camera, a tablet computer camera, a camera on access control equipment and the like, and the electronic equipment can be a mobile phone processor, a tablet computer processor, a background server of an access control system and the like.
By applying the embodiment of the application, if the first face registration image is an image after the face beautifying process, the face beautifying removing process is performed on the face in the image to obtain the second face registration image, that is, the face in the first face registration image is the face after the face beautifying process, and the face in the second face registration image is the face after the face beautifying removing process, the extracted face features include the face features of the face beautifying process and the face features of the face beautifying removing process, so that when the face registration is performed on the face registration image of the first face subjected to the face beautifying process, the face features of the face beautifying process are stored in the face library, and the face features of the face beautifying removing process are also stored, namely the face features in the face library are expanded, and the face features of the face beautifying removing process are provided, so that the accuracy of face verification is improved.
It is noted that relational terms such as first and second, and the like are 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. Moreover, 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 one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for an apparatus, an electronic device, a machine-readable storage medium, a computer program product, a face verification system embodiment, since it is substantially similar to a method embodiment, the description is relatively simple, and reference is made to the part of the description of a method embodiment for relevant points.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the scope of the present application. Any modifications, equivalent substitutions, improvements, etc. that are within the spirit and principles of the present application are intended to be included within the scope of the present application.
Claims (6)
1. A face registration method, the method comprising:
acquiring a first face registration image;
performing face beautifying detection on the first face registration image, and judging whether the face in the first face registration image is subjected to face beautifying treatment or not;
if yes, carrying out face beautifying removal processing on the face to obtain a second face registration image of the face after the face beautifying removal processing;
Extracting face features of the first face registration image and the second face registration image respectively, and storing the extracted face features into a face library;
the face removing processing is performed on the face to obtain a second face registration image of the face after the face removing processing, including:
inputting the first face registration image into a face beauty removal model trained in advance, and performing face beauty removal processing on a face in the first face registration image by using the face beauty removal model to obtain a second face registration image of the face after the face beauty removal processing, wherein the face beauty removal model is a deep network model obtained by training by taking a face image subjected to face beauty processing as a training sample and taking a face image not subjected to face beauty processing as a training target;
the step of performing face beautifying detection on the first face registration image and judging whether the face in the first face registration image is subjected to face beautifying processing comprises the following steps:
inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment.
2. A face verification method, the method comprising:
acquiring a face image to be verified;
extracting face features of the face image to be verified;
matching the extracted face features with each face feature stored in a face library, wherein the face library at least stores face features obtained by extracting face features of a first face registration image, and if face beautifying detection is performed on the first face registration image, face features obtained by extracting face features of a second face registration image are also stored in the face library if face beautifying processing is performed on the first face registration image, and the second face registration image is obtained by performing face beautifying removal processing on the first face registration image; face beautifying detection is carried out on the first face registration image in the following mode: inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment; the first face image is subjected to face beautifying removal processing in the following mode, and the second face registration image is obtained: inputting the first face registration image into a face beauty removal model trained in advance, and performing face beauty removal processing on a face in the first face registration image by using the face beauty removal model to obtain a second face registration image of the face after the face beauty removal processing, wherein the face beauty removal model is a deep network model obtained by training by taking a face image subjected to face beauty processing as a training sample and taking a face image not subjected to face beauty processing as a training target;
And if the face features matched with the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes verification.
3. The method according to claim 2, wherein the acquiring the face image to be verified comprises:
receiving a face image to be verified, which is acquired by a face acquisition unit on access control equipment;
if the face features matched with the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes verification, and then the method further comprises the following steps:
and sending a release notice to the access control equipment to inform the access control equipment of releasing.
4. A face registration apparatus, the apparatus comprising:
the acquisition module is used for acquiring a first face registration image;
the detection module is used for carrying out face beautifying detection on the first face registration image and judging whether the face in the first face registration image is subjected to face beautifying treatment or not;
the beauty Yan Quchu module is used for performing beauty removal processing on the face if the detection result of the detection module is that the face in the first face registration image is subjected to the beauty treatment, so as to obtain a second face registration image of the face after the beauty removal processing;
The feature extraction module is used for extracting the face features of the first face registration image and the second face registration image respectively and storing the extracted face features into a face library;
the beauty Yan Quchu module performs a face beauty removal process on the face to obtain a second face registration image of the face after the face beauty removal process, including:
inputting the first face registration image into a face beauty removal model trained in advance, and performing face beauty removal processing on a face in the first face registration image by using the face beauty removal model to obtain a second face registration image of the face after the face beauty removal processing, wherein the face beauty removal model is a deep network model obtained by training by taking a face image subjected to face beauty processing as a training sample and taking a face image not subjected to face beauty processing as a training target;
the detection module performs face beautifying detection on the first face registration image, and judges whether the face in the first face registration image is subjected to face beautifying processing or not, and the detection module comprises:
inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment.
5. A face verification apparatus, the apparatus comprising:
the acquisition module is used for acquiring the face image to be verified;
the feature extraction module is used for extracting the face features of the face image to be verified;
the matching module is used for matching the extracted face features with each face feature stored in a face library, wherein the face library at least stores face features obtained by extracting face features of a first face registration image, and if face beautifying detection is carried out on the first face registration image, face features obtained by extracting face features of a second face registration image are also stored in the face library if face beautifying processing is carried out on the first face registration image, and the second face registration image is obtained by carrying out face beautifying removal processing on the first face registration image; face beautifying detection is carried out on the first face registration image in the following mode: inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment; the first face image is subjected to face beautifying removal processing in the following mode, and the second face registration image is obtained: inputting the first face registration image into a face beauty removal model trained in advance, and performing face beauty removal processing on a face in the first face registration image by using the face beauty removal model to obtain a second face registration image of the face after the face beauty removal processing, wherein the face beauty removal model is a deep network model obtained by training by taking a face image subjected to face beauty processing as a training sample and taking a face image not subjected to face beauty processing as a training target;
And the verification module is used for determining that the face image to be verified passes verification if the face features obtained by extracting the face features of the face image to be verified exist in the face library.
6. The face verification system is characterized by comprising face acquisition equipment and electronic equipment;
the face acquisition equipment is used for acquiring a face image to be verified and sending the face image to be verified to the electronic equipment;
the electronic equipment is used for receiving the face image to be verified, which is sent by the face acquisition equipment, and extracting face characteristics of the face image to be verified; matching the extracted face features with each face feature stored in a face library, wherein the face library at least stores face features obtained by extracting face features of a first face registration image, and if face beautifying detection is performed on the first face registration image, face features obtained by extracting face features of a second face registration image are also stored in the face library if face beautifying processing is performed on the first face registration image, and the second face registration image is obtained by performing face beautifying removal processing on the first face registration image; face beautifying detection is carried out on the first face registration image in the following mode: inputting the first face registration image into a face liquefaction detector, and predicting an optical flow field of the first face registration image by using the face liquefaction detector to obtain a detection result of whether the face in the first face registration image is subjected to face beautifying treatment; the first face image is subjected to face beautifying removal processing in the following mode, and the second face registration image is obtained: inputting the first face registration image into a face beauty removal model trained in advance, and performing face beauty removal processing on a face in the first face registration image by using the face beauty removal model to obtain a second face registration image of the face after the face beauty removal processing, wherein the face beauty removal model is a deep network model obtained by training by taking a face image subjected to face beauty processing as a training sample and taking a face image not subjected to face beauty processing as a training target; and if the face features matched with the face features obtained by extracting the face features of the face image to be verified exist in the face library, determining that the face image to be verified passes verification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011095448.8A CN112101296B (en) | 2020-10-14 | 2020-10-14 | Face registration method, face verification method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011095448.8A CN112101296B (en) | 2020-10-14 | 2020-10-14 | Face registration method, face verification method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112101296A CN112101296A (en) | 2020-12-18 |
| CN112101296B true CN112101296B (en) | 2024-03-08 |
Family
ID=73783307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011095448.8A Active CN112101296B (en) | 2020-10-14 | 2020-10-14 | Face registration method, face verification method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112101296B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113793433A (en) * | 2021-08-24 | 2021-12-14 | 赵佳 | Household safety intelligent access control system |
| CN113822245B (en) * | 2021-11-22 | 2022-03-04 | 杭州魔点科技有限公司 | Face recognition method, electronic device, and medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013030078A (en) * | 2011-07-29 | 2013-02-07 | Secom Co Ltd | Face image authentication apparatus |
| CN105654033A (en) * | 2015-12-21 | 2016-06-08 | 小米科技有限责任公司 | Face image verification method and device |
| CN109918983A (en) * | 2017-12-12 | 2019-06-21 | 三星电子株式会社 | The device and method that user is registered |
| CN111104825A (en) * | 2018-10-26 | 2020-05-05 | 北京陌陌信息技术有限公司 | Face registry updating method, device, equipment and medium |
| CN111199176A (en) * | 2018-11-20 | 2020-05-26 | 浙江宇视科技有限公司 | Face identity detection method and device |
| CN111598051A (en) * | 2020-06-16 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Face verification method, device and equipment and readable storage medium |
| CN111611873A (en) * | 2020-04-28 | 2020-09-01 | 平安科技（深圳）有限公司 | Face replacement detection method and device, electronic equipment and computer storage medium |
-
2020
- 2020-10-14 CN CN202011095448.8A patent/CN112101296B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013030078A (en) * | 2011-07-29 | 2013-02-07 | Secom Co Ltd | Face image authentication apparatus |
| CN105654033A (en) * | 2015-12-21 | 2016-06-08 | 小米科技有限责任公司 | Face image verification method and device |
| CN109918983A (en) * | 2017-12-12 | 2019-06-21 | 三星电子株式会社 | The device and method that user is registered |
| CN111104825A (en) * | 2018-10-26 | 2020-05-05 | 北京陌陌信息技术有限公司 | Face registry updating method, device, equipment and medium |
| CN111199176A (en) * | 2018-11-20 | 2020-05-26 | 浙江宇视科技有限公司 | Face identity detection method and device |
| CN111611873A (en) * | 2020-04-28 | 2020-09-01 | 平安科技（深圳）有限公司 | Face replacement detection method and device, electronic equipment and computer storage medium |
| CN111598051A (en) * | 2020-06-16 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Face verification method, device and equipment and readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 《Detecting Photoshopped Faces by Scripting Photoshop》;Sheng-Yu Wang et al.;《https://arxiv.org/pdf/1906.05856v2.pdf》;第1-16页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112101296A (en) | 2020-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111340008B (en) | Method and system for generation of counterpatch, training of detection model and defense of counterpatch | |
| US11449971B2 (en) | Method and apparatus with image fusion | |
| WO2021012526A1 (en) | Face recognition model training method, face recognition method and apparatus, device, and storage medium | |
| US10235603B2 (en) | Method, device and computer-readable medium for sensitive picture recognition | |
| WO2019237846A1 (en) | Image processing method and apparatus, face recognition method and apparatus, and computer device | |
| US20170262472A1 (en) | Systems and methods for recognition of faces e.g. from mobile-device-generated images of faces | |
| KR102415509B1 (en) | Face verifying method and apparatus | |
| WO2018156478A1 (en) | Image recognition method and apparatus | |
| WO2016177259A1 (en) | Similar image recognition method and device | |
| CN112052831B (en) | Method, device and computer storage medium for face detection | |
| EP3035240A1 (en) | Authentication apparatus, method, and program | |
| TW202026948A (en) | Methods and devices for biological testing and storage medium thereof | |
| CN111476268A (en) | Method, device, equipment and medium for training reproduction recognition model and image recognition | |
| CN112101296B (en) | Face registration method, face verification method, device and system | |
| EP3437020B1 (en) | Secure storage of fingerprint related elements | |
| CN110532746B (en) | Face checking method, device, server and readable storage medium | |
| CN113642639B (en) | Living body detection method, living body detection device, living body detection equipment and storage medium | |
| KR20150107499A (en) | Object recognition apparatus and control method thereof | |
| CN107247936A (en) | Image-recognizing method and device | |
| CN111353368A (en) | Pan-tilt camera, face feature processing method and device and electronic equipment | |
| CN110059579A (en) | For the method and apparatus of test alive, electronic equipment and storage medium | |
| CN111311550B (en) | Identification method and device for forged certificate and electronic equipment | |
| CN111079587B (en) | Face recognition method and device, computer equipment and readable storage medium | |
| CN114663726A (en) | Training method of target type detection model, target detection method and electronic equipment | |
| CN112560791B (en) | Recognition model training method, recognition method and device and electronic equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |