CN108875654A - A kind of face characteristic acquisition method and device - Google Patents
A kind of face characteristic acquisition method and device Download PDFInfo
- Publication number
- CN108875654A CN108875654A CN201810659963.0A CN201810659963A CN108875654A CN 108875654 A CN108875654 A CN 108875654A CN 201810659963 A CN201810659963 A CN 201810659963A CN 108875654 A CN108875654 A CN 108875654A
- Authority
- CN
- China
- Prior art keywords
- face
- characteristic information
- image
- image set
- similarity value
- 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.)
- Granted
Links
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
- G06V40/168—Feature extraction; Face representation
-
- 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)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of face characteristic acquisition method and device, this method includes but is not limited to following steps：Obtain the image including face；The face characteristic information in described image is detected, described image includes auxiliary information；The determining default characteristic information to match with the face characteristic information in multiple default characteristic informations；Wherein, the corresponding face image set of each default characteristic information；It will be in the corresponding face image set of described image is added to the face characteristic information matches default characteristic information；Wherein, each face image set includes multiple images with different auxiliary informations.The present invention can be realized the training data of the automatic face recognition algorithms for collecting high quality, improve the efficiency of data acquisition, reduce manpower and material resources cost.
Description
Technical field
The present invention relates to computerized information field more particularly to a kind of face characteristic acquisition methods and device.
Background technique
Recognition of face is a kind of biological identification technology for carrying out identification based on facial feature information of people, passes through camera shooting
Machine or camera acquire image or video flowing containing face, and automatic detection and tracking face in the picture, and then to detection
To a series of the relevant technologies for being identified of face.Wherein, face recognition algorithms are the cores in face recognition technology, and people
The accuracy of identification of face recognizer is heavily dependent on training data, and the quality of training data is higher, face recognition algorithms
Accuracy of identification it is higher.However, collecting high quality training data needs to put into a large amount of manpower and material resources cost.
Summary of the invention
The embodiment of the invention provides a kind of face characteristic acquisition method and devices, can collect the image of high quality automatically
As the training data of face recognition algorithms, to reduce manpower and material resources cost.
In a first aspect, this application provides a kind of face characteristic acquisition method, this method includes：
Obtain the image including face；
The face characteristic information in described image is detected, described image includes auxiliary information；
The determining default characteristic information to match with the face characteristic information in multiple default characteristic informations；Wherein,
The corresponding face image set of each default characteristic information；
By the corresponding facial image of described image is added to the face characteristic information matches default characteristic information
In set；Wherein, each face image set includes multiple images with different auxiliary informations.
Second aspect, this application provides face characteristic acquisition device, which includes：
Acquisition unit, for obtaining the image including face；
Detection unit, for detecting the face characteristic information in described image, described image includes auxiliary information；
Matching unit, for the default spy to match with the face characteristic information determining in multiple default characteristic informations
Reference breath；Wherein, the corresponding face image set of each default characteristic information in the multiple default characteristic information；
Adding unit, for described image to be added to the default characteristic information pair to match with the face characteristic information
In the face image set answered；Wherein, each face image set includes multiple images with different auxiliary informations.
The third aspect, this application provides a kind of terminal, which includes processor, memory, input-output system, institute
It states processor, memory, input-output system to be connected with each other, wherein the memory is for storing computer program, the meter
Calculation machine program includes program instruction, and the processor is configured for calling described program instruction, is executed as described in relation to the first aspect
Method.
Fourth aspect, this application provides a kind of system, which includes terminal and camera, the camera and institute
It states and is independently arranged between terminal, the terminal and the camera communicate to connect, and the camera is for acquiring image and by institute
It states image and is sent to the terminal, the terminal is used to execute method as described in relation to the first aspect.
5th aspect, this application provides a kind of computer readable storage medium, the computer storage medium is stored with
Computer program, the computer program include program instruction, and described program instruction makes the processing when being executed by a processor
Device executes method as described in relation to the first aspect.
As can be seen that the application can detect the face characteristic letter for the personnel of being taken from the image including auxiliary information
Breath, then image is added in the corresponding face image set of default characteristic information with after default characteristic information matching.Work as face
When in image collection including multiple images with different auxiliary informations, such face image set can be used as high quality
Face recognition algorithms training data；That is, can obtain being based on angle, light by the Image Acquisition of certain time
According to the different aspects such as, background, age, posture, wearing ornaments the personnel that are taken image as training data, these images are anti-
The abundant in content, quality reflected is high.And compared with different time specifically acquires training data, the time is more saved, it can be effectively
The efficiency for improving collecting training data process, is greatly reduced manpower and material resources cost.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of face characteristic acquisition method provided in an embodiment of the present invention；
Fig. 2 is the flow diagram of another face characteristic acquisition method provided in an embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of face characteristic acquisition device provided in an embodiment of the present invention；
Fig. 4 is a kind of structural schematic diagram of terminal provided in an embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of system provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
Referring to Fig. 1, Fig. 1 is a kind of flow diagram of face characteristic acquisition method provided in an embodiment of the present invention.This
The face method for collecting characteristics mainly is applied to come in terminal having data processing function for example, the terminal by embodiment
It can be Intelligent bracelet, smartwatch, portable digital player, smart phone, palm PC, tablet computer, notebook electricity
Brain, desktop computer, server etc..The face method for collecting characteristics includes but is not limited to following steps：
Step 101, the image including face is obtained.
In the embodiment of the present invention, terminal can be by receiving the image of other equipment transmission, to obtain the figure including face
Picture；Terminal can also shoot to obtain the image for including face by camera.Wherein, camera can be digital camera
Head, simulation camera, charge-coupled device (charge-coupled device, CCD) camera or complementary metal oxide half
Conductor (complementary metal oxide semiconductor, CMOS) camera.Further, camera can be with
It is, such as mobile phone camera integrally formed with terminal, is also possible to independently set in the case where existing and communicating to connect with terminal
It sets, such as remote camera.
It is carried out it should be noted that terminal can open camera after the shooting instruction for receiving the personnel's triggering that is taken
Shooting, to obtain the image for the personnel that are taken.Terminal can also indicate that camera keeps continuing shooting state, when being taken
When personnel are appeared in shooting area, terminal obtains the image of the personnel that are taken.It should be understood that above-mentioned example is only used for illustrating,
It is not specifically to limit.
Step 103, the face characteristic information in described image is detected.
In the embodiment of the present invention, the face characteristic information is that the personnel's face that is taken embodied in described image is special
The relevant information of sign, the face characteristic information can be facial image, face feature vector, facial image set of eigenvectors etc.
Deng being not limited thereto.In the embodiment of the present invention, the image that terminal shoots camera is detected, and is extracted and is clapped
Take the photograph the face characteristic information of personnel.For example, terminal can identify the human face region for the personnel that are taken in the image, and should
Image in human face region carries out feature extraction, obtains the face characteristic information of the personnel that are taken.
Further, described image includes auxiliary information；The auxiliary information may include light, background, facial angle,
One of personnel's posture, wearing ornaments, personnel's age are a variety of；It should be understood that the auxiliary information is for reflecting the people that is taken
The information of many aspects such as angle, illumination, background, age, posture, the wearing ornaments at face position of member, that is to say, that described
Image is the personnel of being taken in the image at moment that is taken, and the auxiliary information of the different images for the personnel that are taken is different.
It should be noted that terminal can detect the figure by preset Feature Selection Model (or feature extraction algorithm)
Face characteristic information as in.This feature, which extracts model, can be neural network model, specifically, including but not limited to convolution mind
Through network (Convolutional Neural Network, CNN) model, residual error network (Residual Networks,
ResNet) model, full convolutional network (Fully Convolutional Networks, FCN) model, multitask cascade
One of which in MNC model, Mask-RCNN model.
Step 105, the determining default feature to match with the face characteristic information is believed in multiple default characteristic informations
Breath.
The corresponding people one by one of each default characteristic information in the embodiment of the present invention, in the multiple default characteristic information
Face image set, each face image set include multiple images with different auxiliary informations of the personnel that are taken.Institute
Stating default characteristic information is the pre-set face characteristic information being acquired to the personnel of being taken, such as default feature
Be taken the resident identification card of personnel, the face characteristic information for the acquisitions such as photo or passport of opening an account according to information.Possible
It include the corresponding image of default characteristic information of the corresponding personnel that are taken in embodiment, in the face image set.
In a specific embodiment, terminal can detect the face characteristic information and the multiple default feature respectively
The similarity value between each default characteristic information in information, determines maximum similarity value；Judge described maximum
Whether similarity value is greater than or equal to the first preset threshold；It is greater than or equal to the first preset threshold in the maximum similarity value
In the case where, using the maximum similarity value corresponding default characteristic information as matching with the face characteristic information
Default characteristic information.
For example, the similarity value between face characteristic information and multiple default characteristic informations is respectively：3%, 64%,
23%, 89%, 8% ... terminal determines therein 89% for maximum similarity value, and by 89% and first preset threshold
85% is compared, so that it is determined that the similarity value meets the first preset threshold.Terminal further detects described maximum
The corresponding default face information of similarity, the default spy which is matched as the face characteristic information
Reference breath.
In a specific embodiment, the face characteristic information that terminal can will test is input to human face recognition model
In, the human face recognition model is after being trained in advance by the corresponding default characteristic information of multiple and different personnel that are taken
Model.Terminal can identify to obtain the default feature letter that the face characteristic information matches based on the human face recognition model
Breath.It should be understood that the example in above-mentioned specific embodiment is only used for illustrating, the matching treatment of other way can also be, herein
It is not construed as limiting.
Step 107, described image is added to the face characteristic information matches default characteristic information is corresponding
In face image set.
In the embodiment of the present invention, terminal is in the determining default characteristic information to match with collected face characteristic information
Afterwards, by the corresponding face image set of described image is added to the face characteristic information matches default characteristic information
In.It should be noted that the face image set is combined into the set of the image for the personnel that are individually taken, meanwhile, the face figure
It is the set that can be used for the training data for the personnel that are individually taken of model training that image set, which closes,.
It is worth noting that in model training, if as the image in the face image set of training data angle,
Illumination, background, age, posture, wearing ornaments etc. are excessively single, then the model performance trained is bad, and precision is not high.It changes
Sentence is talked about, in multiple and different images of the same personnel that are taken, in angle, illumination, background, age, posture, wearing ornaments etc.
It covers more complete, then it represents that these images are higher as the quality of training data.Further, the embodiment of the present invention can pass through one section
The image acquisition and processing of time (such as one month, a season etc.), obtains the multiple images of the same personnel that are taken, described more
A image is taken in the corresponding face image set of personnel described in being all stored in, multiple figures in the face image set
As being respectively provided with different auxiliary informations.For example, a user can be acquired image zooming-out face characteristic letter by camera daily
Breath, but the user is daily in the image of camera-shot, the angle, illumination, background, age, the posture, jewellery that are presented
Wearing etc. is all different, so that the image in the corresponding face image set of the user is different.These images are being used for model
When training, it is capable of providing the higher training dataset of quality, training obtains the better model of performance.
One possible application scenarios of the embodiment of the present invention are described below with reference in conjunction with above method process, in order into
One step understands inventive concept of the invention.
In a possible application scenarios, one or more places are provided with camera, and it is big that the place can be bank
The biggish positions of flows of the people such as the Room, government's working hall.The camera and terminal communicate to connect, and the terminal is taken the photograph by receiving
The video image sent as hair carries out face characteristic information acquisition.Terminal, which collects, appears in being taken for camera shooting area
After the image of personnel, the face characteristic information in described image is detected, and by the face characteristic information and be stored in advance in terminal
Multiple default characteristic informations in memory are matched, and the default feature to match with the face characteristic information letter is obtained
Breath.Then the acquired image is added in the corresponding face image set of the default characteristic information by terminal.
As can be seen that through the embodiment of the present invention, can shoot the personnel of being taken appeared in camera is included
The image of the personnel's face that is taken, and detect with match described in be taken the face characteristic information of personnel, described image is added
Add in corresponding face image set, it is not only more convenient practical, improve data acquisition efficiency, can also accurately by by
The image of shooting personnel is added in corresponding face image set, convenient subsequently to use.And the figure of these personnel that are taken
The different aspects such as various angles, illumination, background, age, posture, wearing ornaments of the picture including the personnel's face that is taken
Auxiliary information, face is abundant in content, the quality of data is high, effectively improves the efficiency of collecting training data process, is greatly reduced
Manpower and material resources cost.
Based on the same inventive concept, the embodiment of the invention also provides a kind of signals of the process of face characteristic acquisition method
Figure, referring to fig. 2.The face method for collecting characteristics includes but is not limited to following steps：
Step 201, the image for obtaining camera shooting detects the face characteristic in described image by Feature Selection Model
Information.
In a specific embodiment, terminal detects that the image middle finger of camera shooting is shown with the case where being taken personnel
Under, start to detect the face characteristic information in described image.For example, the image that camera takes remains static, it is described
Stationary state indicates do not have the object or person of Large Amplitude Motion in image.When terminal detects that described image is in nonstatic shape
State starts the face characteristic information in detection image.
In a specific embodiment, the method can also include fuzzy judgement step.Specifically, terminal can identify
Human face region in described image, and the area image is subjected to fog-level detection, when fog-level is higher than preset standard, eventually
It holds the processing terminated to described image and the image for obtaining the subsequent shooting of camera is handled.
The specific implementation process of the step 201 can refer to the associated description of Fig. 1 embodiment step 101 and step 103, this
In repeat no more.
Step 203, it is based on the face characteristic information, it is determining in multiple default characteristic informations to believe with the face characteristic
The matched default characteristic information of manner of breathing.The specific implementation process of the step 203 can refer to the correlation of Fig. 1 embodiment step 105
Description, which is not described herein again.
Step 205, described image is added to the face characteristic information matches default characteristic information is corresponding
In face image set.The specific implementation process of the step 205 can refer to the associated description of Fig. 1 embodiment step 107, here
It repeats no more.
Step 207, data cleansing is carried out to the face image set, obtains the face image set that can be used for model training
It closes.
In the embodiment of the present invention, the data cleansing is for finding and correcting in face image set identifiable quality not
High or wrong face characteristic information.For example, terminal shoots the face characteristic that image detects by camera
There is no registrations in the terminal to obtain default characteristic information by the corresponding personnel that are taken of information, i.e., does not have in multiple default characteristic informations
There is the default characteristic information of the personnel that are taken.After terminal acquires the face characteristic information of the personnel that are taken, terminal is true
The default face information that the face characteristic information of the fixed personnel that are taken matches, and described image is added to the default people
In the corresponding face image set of face information.It is understood that the personnel of being taken not are the face image set pair
The user answered, the i.e. image of the personnel that are taken not should be in the face image set.So data cleansing is in the example
In for deleting the image of the personnel that are taken from the face image set.It should be understood that above-mentioned example is only for example, it is not
It is specific to limit.
In a specific embodiment, terminal, which carries out data cleansing to the face image set, can be：Terminal is in institute
It states and determines target image in the multiple images of face image set；Face characteristic information and the institute of the target image are detected respectively
State the similarity value between the face characteristic information of other each images in face image set in addition to the target image；System
Count other each figures in the face characteristic information and the face image set of the target image in addition to the target image
Quantity of the similarity value less than the image of the second preset threshold in similarity value between the face characteristic information of picture；In the number
In the case that the ratio that amount accounts for the quantity of image in the face image set is greater than or equal to third predetermined threshold value, from the people
The target image is deleted in face image set.
For example, the total amount of image is 1000 in face image set, face is based between terminal statistics and target image
The quantity of face characteristic information of the similarity of characteristic information less than 95% is 50, which accounts for the 5% of total amount, so
Ratio is more than third predetermined threshold value 3%, and terminal deletes target face characteristic information.It should be noted that in possible implementation
Second preset threshold described in example can be equal to the first preset threshold in abovementioned steps 105.
It should also be noted that, the data cleansing step in above-mentioned specific embodiment is in a face image set
The treatment process made of an image.In further implementation process, terminal can be to the every of each face image set
A image carries out the data cleansing of above-mentioned specific embodiment such as and handles.
In a specific embodiment, terminal, which carries out data cleansing to the face image set, can also be：Terminal point
It does not detect similar between the face characteristic information of each image and corresponding default characteristic information in the face image set
The whole similarity values found out are obtained average similarity value as average calculating operation by angle value.Average similarity value is subtracted one by terminal
The value obtained after preset value deletes similarity value less than the 4th threshold value as the 4th preset threshold, and from face image set
Image.
For example, the total amount of face characteristic information is 600 in face image set, terminal will be each in face image set
Similarity value between the face characteristic information of image and corresponding default characteristic information is averaged, and obtains average similarity value and is
96%.Terminal is used as the 4th preset threshold for 96% or less 3 percentage point, deletes and the phase between corresponding default characteristic information
Like angle value image corresponding less than 93% face characteristic information.
It should be understood that the example of above-mentioned specific embodiment is only used for illustrating, it is not specifically to limit.
It is worth noting that terminal can execute the processing of step 201 to step 205 within a preset time, it is pre- when reaching
If terminal executes step 207 again after the time.For example, terminal executes step 201 to step 205, to acquire enough in one month
The face characteristic information of quantity executes step 207 when terminal detects that acquisition time reaches one month, clear to carry out data
It washes.
Step 209, the Feature Selection Model is instructed according to the face image set that can be used for model training
Practice, to update the model parameter of the Feature Selection Model.
In the embodiment of the present invention, terminal is by the face image set that can be used for model training Jing Guo data cleansing to described
Feature Selection Model is trained, to update the model parameter of the Feature Selection Model, so as to extract more accurate people
Face characteristic information.It should be noted that the face image set that can be used for model training can be also used for other models
Training, such as terminal can be used for the face image set of model training according to and be trained to human face recognition model, can be with
The model parameter of the human face recognition model is updated, so that the human face recognition model can more accurately identify people
Face.
As can be seen that through the embodiment of the present invention, the personnel that are taken that can be will appear in camera carry out image and adopt
Image, is then added in corresponding face image set, more by collection, and the face characteristic information for the personnel that are taken described in detection
Add convenient practical, the efficiency of raising data acquisition.Terminal can also carry out data cleansing to face image set, obtain more quasi-
Face image set that is true and being more suitable for model training.And the face image set can be also used for Feature Selection Model
The performance that this feature extracts model can be continuously improved in training by way of iteration.
Based on the same inventive concept, described referring to Fig. 3 the embodiment of the invention provides a kind of face characteristic acquisition device
Terminal include at least acquiring unit 301, detection unit 303, determination unit 305, adding unit 307, the terminal for realizing
Face characteristic acquisition method described in Fig. 1, Fig. 2 embodiment of the method.
Acquiring unit 301, for obtaining the image including face；
Detection unit 303, for detecting the face characteristic information in described image, described image includes auxiliary information；
Determination unit 305, for determined in multiple default characteristic informations match with the face characteristic information it is pre-
If characteristic information；Wherein, the corresponding face image set of each default characteristic information in the multiple default characteristic information；
Adding unit 307 is believed for described image to be added to the default feature to match with the face characteristic information
It ceases in corresponding face image set；Wherein, each face image set includes multiple images with different auxiliary informations.
Specifically, the determination unit 305 is used for：The face characteristic information and the multiple default feature are detected respectively
The similarity value between each default characteristic information in information, determines maximum similarity value；Judge described maximum
Whether similarity value is greater than or equal to the first preset threshold；It is greater than or equal to the first preset threshold in the maximum similarity value
In the case where, using the maximum similarity value corresponding default characteristic information as matching with the face characteristic information
Default characteristic information.
Optionally, the auxiliary information includes light, background, facial angle, personnel's posture, wearing ornaments, personnel's age
One of or it is a variety of；Described device further includes：Data cleansing unit, for after reaching preset time, to the face figure
Image set, which closes, carries out data cleansing, obtains the face image set that can be used for model training.
Specifically, the data cleansing unit is used for：Target figure is determined in the multiple images of the face image set
Picture；Its in the face characteristic information and the face image set of the target image in addition to the target image is detected respectively
Similarity value between the face characteristic information of its each image；Count the target image face characteristic information and the people
It is similar in similarity value between the face characteristic information of other each images in face image set in addition to the target image
Quantity of the angle value less than the image of the second preset threshold；The ratio of the quantity of image in the face image set is accounted in the quantity
In the case that example is greater than or equal to third predetermined threshold value, the target image is deleted from the face image set.
Optionally, it is special to be specifically used for the face detected in described image by Feature Selection Model for the detection unit 303
Reference breath；Described device further includes updating unit, for can be used for the face image set of model training according to described
Feature Selection Model is trained, to update the model parameter of the Feature Selection Model.
It should be noted that by earlier figures 1, the detailed description of Fig. 2 embodiment of the method, those skilled in the art can understand
The implementation method of each unit that face characteristic acquisition device is included is known on ground, so in order to illustrate the succinct of book, herein not
It repeats again.
Based on the same inventive concept, the embodiment of the invention provides a kind of terminals, referring to fig. 4, the terminal for realizing
Face characteristic acquisition method described in Fig. 1, Fig. 2 embodiment of the method.As shown in figure 4, the terminal may include：Processor 401,
Memory 402, input-output system 403.These components can be communicated on one or more communication bus 404.The end
End can also include communication module 405, power module 406.
Alleged processor 401 can be central processing unit (Central Processing Unit, CPU), the processor
It can also be other general processors, digital signal processor (Digital Signal Processor, DSP), dedicated integrated
Circuit (Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 402 may include read-only memory and random access memory, and provide instruction to processor 401
And data.The a part of of memory 402 can also include nonvolatile RAM.In addition, memory 402 can be with
The information of storage device type.
The input-output system 403 is mainly used for receiving user instructions and shooting image.In the specific implementation, input is defeated
System may include out：Camera controller 4031.Wherein, each controller can be with corresponding peripheral equipment (camera
4032) it couples.It should be noted that input-output system 403 can also include other I/O peripheral hardwares.
Wherein, the camera 4033 is the executing agency of the terminal.Specifically, camera 4033 is used for shooting area
Domain is shot to obtain image.
The communication module 405 is mainly used for being communicated with server；The power module 406 is mainly used for as equipment
In other devices stable power supply is provided.
In the embodiment of the present invention, the processor 401 is held for calling the instruction stored in the memory 402
Row following steps：
Obtain the image including face；
The face characteristic information in described image is detected, described image includes auxiliary information；
The determining default characteristic information to match with the face characteristic information in multiple default characteristic informations；Wherein,
The corresponding face image set of each default characteristic information；
By the corresponding facial image of described image is added to the face characteristic information matches default characteristic information
In set；Wherein, each face image set includes multiple images with different auxiliary informations.
In one embodiment, the processor 401 can specifically execute following steps：It is special that the face is detected respectively
The similarity value between each default characteristic information in reference breath and the multiple default characteristic information, determines maximum
Similarity value；Judge whether the maximum similarity value is greater than or equal to the first preset threshold；In the maximum similarity
In the case that value is greater than or equal to the first preset threshold, using the corresponding default characteristic information of the maximum similarity value as with
The default characteristic information that the face characteristic information matches.
In one embodiment, the auxiliary information includes light, background, facial angle, personnel's posture, jewellery pendant
It wears, one of personnel's age or a variety of；The processor 401 may call upon the instruction stored in the memory 402 and hold
Row following steps：After reaching preset time, data cleansing is carried out to the face image set, obtains can be used for model training
Face image set.
In one embodiment, the processor 401 can specifically execute following steps：In the face image set
Multiple images in determine target image；The face characteristic information and the face image set of the target image are detected respectively
In other each images in addition to the target image face characteristic information between similarity value；Count the target image
Face characteristic information and the face image set in the face characteristics of other each images in addition to the target image believe
Quantity of the similarity value less than the image of the second preset threshold in similarity value between breath；The face figure is accounted in the quantity
In the case that the ratio of the quantity of image is greater than or equal to third predetermined threshold value in image set conjunction, deleted from the face image set
Except the target image.
In one embodiment, the processor 401 can specifically execute following steps：It is examined by Feature Selection Model
Survey the face characteristic information in described image.The processor may call upon the instruction execution stored in the memory 402：
The Feature Selection Model is trained according to the face image set that can be used for model training, to update the feature
Extract the model parameter of model.
It should be noted that those skilled in the art can be clear by earlier figures 1 or the detailed description of Fig. 2 embodiment of the method
The implementation method for each function element that terminal is included is known to Chu, so details are not described herein in order to illustrate the succinct of book.
Based on the same inventive concept, the embodiment of the invention provides a kind of system, which includes terminal and camera,
The camera is for acquiring image and described image being sent to the terminal, and the terminal is for executing following steps：
Obtain the image including face；
The face characteristic information in described image is detected, described image includes auxiliary information；
The determining default characteristic information to match with the face characteristic information in multiple default characteristic informations；Wherein,
The corresponding face image set of each default characteristic information；
By the corresponding facial image of described image is added to the face characteristic information matches default characteristic information
In set；Wherein, each face image set includes multiple images with different auxiliary informations.
Wherein, the camera is the remote camera with communication function, between the camera and the terminal solely
It erects and sets, the camera and the terminal communicate to connect.It is worth noting that can be between the camera and the terminal
It is communicated by wired mode or wireless mode.Wherein, wired mode includes but is not limited to：RS232 mode, the side RS485
Formula, cable mode, copper cable mode etc..Wireless mode includes but is not limited to：Cellular based communication mode, equipment to DeviceMode or
The other wireless communication modes of person.Cellular based communication mode can be using based on global system for mobile communications (Global System
For Mobile, CommunicationGSM), CDMA (Code Division Multiple Access, CDMA) is wide
Band CDMA (Wideband Code Division Multiple Access, WCDMA), long term evolution (Long Term
Evolution, LTE) or other standards, it is not especially limited herein.Equipment is to equipment (Device to Device) side
Formula can be not especially limited device talk mode using WIFI, Zigbee, bluetooth or other equipment herein.
And the camera can be digital camera, simulation camera, charge-coupled device (charge-coupled
Device, CCD) camera or complementary metal oxide semiconductor (complementary metal oxide
Semiconductor, CMOS) camera, it is not especially limited herein.
It should be noted that those skilled in the art can be clear by earlier figures 1 or the detailed description of Fig. 2 embodiment of the method
The function of each device end in the system is known to Chu, so details are not described herein in order to illustrate the succinct of book.
Based on the same inventive concept, a kind of computer readable storage medium, institute are provided in another embodiment of the invention
Stating computer-readable recording medium storage has computer program, and the computer program includes program instruction, described program instruction
Method described in aforementioned either method embodiment is realized when being executed by processor.
The computer readable storage medium can be the internal storage unit of terminal described in aforementioned any embodiment, example
Such as the hard disk or memory of terminal.The computer readable storage medium is also possible to the External memory equipment of the terminal, such as
The plug-in type hard disk being equipped in the terminal, intelligent memory card (Smart Media Card, SMC), secure digital (Secure
Digital, SD) card, flash card (Flash Card) etc..Further, the computer readable storage medium can also be wrapped both
The internal storage unit for including the terminal also includes External memory equipment.The computer readable storage medium is described for storing
Other programs and data needed for computer program and the terminal.The computer readable storage medium can be also used for temporarily
When store the data that has exported or will export.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and method and step, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
It is apparent to those skilled in the art that for convenience of description and succinctly, the list of foregoing description
Member, the specific work process of equipment, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed unit, device and method, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.In addition, shown or beg for
Opinion mutual coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING of device or unit
Or communication connection, it is also possible to electricity, mechanical or other form connections.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.Some or all of unit therein can be selected to realize the embodiment of the present invention according to the actual needs
Purpose.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, is also possible to two or more units and is integrated in one unit.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the present invention
Portion or part steps.And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of face characteristic acquisition method, which is characterized in that including：
Obtain the image including face；
The face characteristic information in described image is detected, described image includes auxiliary information；
The determining default characteristic information to match with the face characteristic information in multiple default characteristic informations；Wherein, each
Default characteristic information corresponds to a face image set；
By the corresponding face image set of described image is added to the face characteristic information matches default characteristic information
In；Wherein, each face image set includes multiple images with different auxiliary informations.
2. the method according to claim 1, wherein the determining and people in multiple default characteristic informations
The default characteristic information that face characteristic information matches, including：
It is detected between each default characteristic information in the face characteristic information and the multiple default characteristic information respectively
Similarity value determines maximum similarity value；
Judge whether the maximum similarity value is greater than or equal to the first preset threshold；
In the case where the maximum similarity value is greater than or equal to the first preset threshold, by the maximum similarity value pair
The default characteristic information answered is as the default characteristic information to match with the face characteristic information.
3. the method according to claim 1, wherein the auxiliary information include light, background, facial angle,
One of personnel's posture, wearing ornaments, personnel's age are a variety of；
It is described by the corresponding face of described image is added to the face characteristic information matches default characteristic information
After in image collection, the method also includes：
After reaching preset time, data cleansing is carried out to the face image set, obtains the face that can be used for model training
Image collection.
4. according to the method described in claim 3, it is characterized in that, it is described to the face image set carry out data cleansing,
Including：
Target image is determined in the multiple images of the face image set；
It is detected in the face characteristic information and the face image set of the target image respectively in addition to the target image
Similarity value between the face characteristic information of other each images；
Count the face characteristic information of the target image with it is other in addition to the target image in the face image set
Quantity of the similarity value less than the image of the second preset threshold in similarity value between the face characteristic information of each image；
It is greater than or equal to the feelings of third predetermined threshold value in the ratio that the quantity accounts for the quantity of image in the face image set
Under condition, the target image is deleted from the face image set.
5. the method according to claim 3 or 4, which is characterized in that
Face characteristic information in the detection described image, including：The people in described image is detected by Feature Selection Model
Face characteristic information；
It is described that data cleansing is carried out to the face image set, it obtains can be used for after the face image set of model training,
The method also includes：The Feature Selection Model is instructed according to the face image set that can be used for model training
Practice, to update the model parameter of the Feature Selection Model.
6. a kind of face characteristic acquisition device, which is characterized in that including：
Acquiring unit, for obtaining the image including face；
Detection unit, for detecting the face characteristic information in described image, described image includes auxiliary information；
Determination unit is believed for the default feature to match with the face characteristic information determining in multiple default characteristic informations
Breath；Wherein, the corresponding face image set of each default characteristic information in the multiple default characteristic information；
Adding unit, for described image is added to the face characteristic information matches default characteristic information is corresponding
In face image set；Wherein, each face image set includes multiple images with different auxiliary informations.
7. device according to claim 6, which is characterized in that the determination unit is specifically used for：
It is detected between each default characteristic information in the face characteristic information and the multiple default characteristic information respectively
Similarity value determines maximum similarity value；
Judge whether the maximum similarity value is greater than or equal to the first preset threshold；
In the case where the maximum similarity value is greater than or equal to the first preset threshold, by the maximum similarity value pair
The default characteristic information answered is as the default characteristic information to match with the face characteristic information.
8. device according to claim 6, which is characterized in that the auxiliary information include light, background, facial angle,
One of personnel's posture, wearing ornaments, personnel's age are a variety of；
Described device further includes：Data cleansing unit, for being counted to the face image set after reaching preset time
According to cleaning, the face image set that can be used for model training is obtained.
9. device according to claim 8, which is characterized in that the data cleansing unit is specifically used for：
Target image is determined in the multiple images of the face image set；
It is detected in the face characteristic information and the face image set of the target image respectively in addition to the target image
Similarity value between the face characteristic information of other each images；
Count the face characteristic information of the target image with it is other in addition to the target image in the face image set
Quantity of the similarity value less than the image of the second preset threshold in similarity value between the face characteristic information of each image；
It is greater than or equal to the feelings of third predetermined threshold value in the ratio that the quantity accounts for the quantity of image in the face image set
Under condition, the target image is deleted from the face image set.
10. device according to claim 8 or claim 9, which is characterized in that
The detection unit is specifically used for detecting the face characteristic information in described image by Feature Selection Model；
Described device further includes updating unit, for can be used for the face image set of model training according to the feature
It extracts model to be trained, to update the model parameter of the Feature Selection Model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810659963.0A CN108875654B (en) | 2018-06-25 | 2018-06-25 | Face feature acquisition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810659963.0A CN108875654B (en) | 2018-06-25 | 2018-06-25 | Face feature acquisition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108875654A true CN108875654A (en) | 2018-11-23 |
| CN108875654B CN108875654B (en) | 2021-03-05 |
Family
ID=64295603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810659963.0A Active CN108875654B (en) | 2018-06-25 | 2018-06-25 | Face feature acquisition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108875654B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110427912A (en) * | 2019-08-12 | 2019-11-08 | 深圳市捷顺科技实业股份有限公司 | A kind of method for detecting human face and its relevant apparatus based on deep learning |
| CN111310580A (en) * | 2020-01-19 | 2020-06-19 | 四川联众竞达科技有限公司 | Face recognition method under non-matching state |
| CN111382647A (en) * | 2018-12-29 | 2020-07-07 | 广州市百果园信息技术有限公司 | Picture processing method, device, equipment and storage medium |
| CN111710085A (en) * | 2020-05-27 | 2020-09-25 | 南京金陵塑胶化工有限公司 | Production workshop safety management method and system based on facial recognition |
| WO2021098801A1 (en) * | 2019-11-20 | 2021-05-27 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Data cleaning device, data cleaning method and face verification method |
Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101396277A (en) * | 2007-09-26 | 2009-04-01 | 中国科学院声学研究所 | Ultrasonics face recognition method and device |
| CN103702126A (en) * | 2013-12-10 | 2014-04-02 | 清华大学深圳研究生院 | Parallel encoding optimization method based on standard video HEVC (High Efficiency Video Coding) |
| CN104182734A (en) * | 2014-08-18 | 2014-12-03 | 桂林电子科技大学 | Linear-regression based classification (LRC) and collaborative representation based two-stage face identification method |
| CN104700076A (en) * | 2015-02-13 | 2015-06-10 | 电子科技大学 | Face image virtual sample generating method |
| CN105184238A (en) * | 2015-08-26 | 2015-12-23 | 广西小草信息产业有限责任公司 | Human face recognition method and system |
| CN105513368A (en) * | 2015-11-26 | 2016-04-20 | 银江股份有限公司 | Uncertain information-based method for screening vehicles with false license plates |
| CN105631404A (en) * | 2015-12-17 | 2016-06-01 | 小米科技有限责任公司 | Method and device for clustering pictures |
| CN106204779A (en) * | 2016-06-30 | 2016-12-07 | 陕西师范大学 | The check class attendance method learnt based on plurality of human faces data collection strategy and the degree of depth |
| CN106295482A (en) * | 2015-06-11 | 2017-01-04 | 中国移动（深圳）有限公司 | The update method of a kind of face database and device |
| CN106529593A (en) * | 2016-11-08 | 2017-03-22 | 广东诚泰交通科技发展有限公司 | Pavement disease detection method and system |
| CN106650804A (en) * | 2016-12-13 | 2017-05-10 | 深圳云天励飞技术有限公司 | Facial sample cleaning method and system based on deep learning features |
| CN107292252A (en) * | 2017-06-09 | 2017-10-24 | 南京华捷艾米软件科技有限公司 | A kind of personal identification method of autonomous learning |
| CN107301578A (en) * | 2016-04-15 | 2017-10-27 | 上海新飞凡电子商务有限公司 | Obtain and recognize the method and its device of customer information |
| CN107423606A (en) * | 2017-08-01 | 2017-12-01 | 黄河科技学院 | A kind of identification system based on fuzzy control theory |
| CN107491685A (en) * | 2017-09-27 | 2017-12-19 | 维沃移动通信有限公司 | A kind of face identification method and mobile terminal |
| CN107563897A (en) * | 2017-09-08 | 2018-01-09 | 廖海斌 | Based on face matching famous person pursue a goal with determination recommendation and social networks method for building up and system |
| CN107679546A (en) * | 2017-08-17 | 2018-02-09 | 平安科技（深圳）有限公司 | Face image data acquisition method, device, terminal device and storage medium |
| CN107909104A (en) * | 2017-11-13 | 2018-04-13 | 腾讯数码（天津）有限公司 | The face cluster method, apparatus and storage medium of a kind of picture |
| CN107944020A (en) * | 2017-12-11 | 2018-04-20 | 深圳云天励飞技术有限公司 | Facial image lookup method and device, computer installation and storage medium |
| CN108052925A (en) * | 2017-12-28 | 2018-05-18 | 江西高创保安服务技术有限公司 | A kind of cell personnel archives intelligent management |
-
2018
- 2018-06-25 CN CN201810659963.0A patent/CN108875654B/en active Active
Patent Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101396277A (en) * | 2007-09-26 | 2009-04-01 | 中国科学院声学研究所 | Ultrasonics face recognition method and device |
| CN103702126A (en) * | 2013-12-10 | 2014-04-02 | 清华大学深圳研究生院 | Parallel encoding optimization method based on standard video HEVC (High Efficiency Video Coding) |
| CN104182734A (en) * | 2014-08-18 | 2014-12-03 | 桂林电子科技大学 | Linear-regression based classification (LRC) and collaborative representation based two-stage face identification method |
| CN104700076A (en) * | 2015-02-13 | 2015-06-10 | 电子科技大学 | Face image virtual sample generating method |
| CN106295482A (en) * | 2015-06-11 | 2017-01-04 | 中国移动（深圳）有限公司 | The update method of a kind of face database and device |
| CN105184238A (en) * | 2015-08-26 | 2015-12-23 | 广西小草信息产业有限责任公司 | Human face recognition method and system |
| CN105513368A (en) * | 2015-11-26 | 2016-04-20 | 银江股份有限公司 | Uncertain information-based method for screening vehicles with false license plates |
| CN105631404A (en) * | 2015-12-17 | 2016-06-01 | 小米科技有限责任公司 | Method and device for clustering pictures |
| CN107301578A (en) * | 2016-04-15 | 2017-10-27 | 上海新飞凡电子商务有限公司 | Obtain and recognize the method and its device of customer information |
| CN106204779A (en) * | 2016-06-30 | 2016-12-07 | 陕西师范大学 | The check class attendance method learnt based on plurality of human faces data collection strategy and the degree of depth |
| CN106529593A (en) * | 2016-11-08 | 2017-03-22 | 广东诚泰交通科技发展有限公司 | Pavement disease detection method and system |
| CN106650804A (en) * | 2016-12-13 | 2017-05-10 | 深圳云天励飞技术有限公司 | Facial sample cleaning method and system based on deep learning features |
| CN107292252A (en) * | 2017-06-09 | 2017-10-24 | 南京华捷艾米软件科技有限公司 | A kind of personal identification method of autonomous learning |
| CN107423606A (en) * | 2017-08-01 | 2017-12-01 | 黄河科技学院 | A kind of identification system based on fuzzy control theory |
| CN107679546A (en) * | 2017-08-17 | 2018-02-09 | 平安科技（深圳）有限公司 | Face image data acquisition method, device, terminal device and storage medium |
| CN107563897A (en) * | 2017-09-08 | 2018-01-09 | 廖海斌 | Based on face matching famous person pursue a goal with determination recommendation and social networks method for building up and system |
| CN107491685A (en) * | 2017-09-27 | 2017-12-19 | 维沃移动通信有限公司 | A kind of face identification method and mobile terminal |
| CN107909104A (en) * | 2017-11-13 | 2018-04-13 | 腾讯数码（天津）有限公司 | The face cluster method, apparatus and storage medium of a kind of picture |
| CN107944020A (en) * | 2017-12-11 | 2018-04-20 | 深圳云天励飞技术有限公司 | Facial image lookup method and device, computer installation and storage medium |
| CN108052925A (en) * | 2017-12-28 | 2018-05-18 | 江西高创保安服务技术有限公司 | A kind of cell personnel archives intelligent management |
Non-Patent Citations (2)
| Title |
|---|
| 周凯汀等: ""基于改进ORB特征的多姿态人脸识别"", 《计算机辅助设计与图形学学报》 * |
| 李文勇等: ""基于机器视觉的作物多姿态害虫特征提取与分类方法"", 《农业工程学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382647A (en) * | 2018-12-29 | 2020-07-07 | 广州市百果园信息技术有限公司 | Picture processing method, device, equipment and storage medium |
| CN111382647B (en) * | 2018-12-29 | 2021-07-30 | 广州市百果园信息技术有限公司 | Picture processing method, device, equipment and storage medium |
| CN110427912A (en) * | 2019-08-12 | 2019-11-08 | 深圳市捷顺科技实业股份有限公司 | A kind of method for detecting human face and its relevant apparatus based on deep learning |
| WO2021098801A1 (en) * | 2019-11-20 | 2021-05-27 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Data cleaning device, data cleaning method and face verification method |
| CN111310580A (en) * | 2020-01-19 | 2020-06-19 | 四川联众竞达科技有限公司 | Face recognition method under non-matching state |
| CN111710085A (en) * | 2020-05-27 | 2020-09-25 | 南京金陵塑胶化工有限公司 | Production workshop safety management method and system based on facial recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108875654B (en) | 2021-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108875654A (en) | A kind of face characteristic acquisition method and device | |
| CN107679448B (en) | Eyeball action-analysing method, device and storage medium | |
| CN108197250B (en) | Picture retrieval method, electronic equipment and storage medium | |
| US9183431B2 (en) | Apparatus and method for providing activity recognition based application service | |
| CN110113523A (en) | Intelligent photographing method, device, computer equipment and storage medium | |
| CN107679449B (en) | Lip motion method for catching, device and storage medium | |
| CN110633004B (en) | Interaction method, device and system based on human body posture estimation | |
| US10650234B2 (en) | Eyeball movement capturing method and device, and storage medium | |
| CN110879995A (en) | Target object detection method and device, storage medium and electronic device | |
| CN109753868A (en) | Appraisal procedure and device, the Intelligent bracelet of athletic performance | |
| CN109147007A (en) | Paster loading method, device, terminal and computer readable storage medium | |
| CN110442742A (en) | Retrieve method and device, processor, electronic equipment and the storage medium of image | |
| CN110443224A (en) | Page turning detection method, device, electronic equipment and storage medium | |
| CN108171208A (en) | Information acquisition method and device | |
| CN110427108A (en) | Photographic method and Related product based on eyeball tracking | |
| CN106075913A (en) | A kind of information processing method and terminal | |
| CN107613550A (en) | Solve lock control method and Related product | |
| CN111857334A (en) | Human body gesture letter recognition method and device, computer equipment and storage medium | |
| CN111062362B (en) | Face living body detection model, method, device, equipment and storage medium | |
| CN114465678A (en) | Complex activity WIFI perception method based on deep learning | |
| CN109740418B (en) | Yoga action identification method based on multiple acceleration sensors | |
| CN107479715A (en) | The method and apparatus that virtual reality interaction is realized using gesture control | |
| CN107402787A (en) | Iris recognition pattern open method and Related product | |
| CN113657154A (en) | Living body detection method, living body detection device, electronic device, and storage medium | |
| CN113627334A (en) | Object behavior identification 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |