CN109840477A - Face identification method and device are blocked based on eigentransformation - Google Patents
Face identification method and device are blocked based on eigentransformation Download PDFInfo
- Publication number
- CN109840477A CN109840477A CN201910006884.4A CN201910006884A CN109840477A CN 109840477 A CN109840477 A CN 109840477A CN 201910006884 A CN201910006884 A CN 201910006884A CN 109840477 A CN109840477 A CN 109840477A
- Authority
- CN
- China
- Prior art keywords
- feature
- neural networks
- convolutional neural
- sample image
- exposure mask
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the present invention, which provides, a kind of is blocked face identification method and device based on eigentransformation, the described method includes: target facial image is input to preset convolutional neural networks model, the characteristic pattern of the target facial image is obtained from the last one convolutional layer of the convolutional neural networks model；Calculate point plain product of the characteristic pattern with the feature exposure mask obtained in advance；Described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the feature for identification of the target facial image.It is provided in an embodiment of the present invention that face identification method and device are blocked based on eigentransformation, to the eigentransformation mode for being used addition feature exposure mask by the image for blocking face, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is easier, it is shorter to calculate the time, network structure is simpler, and improves by the accuracy of identification for blocking face.
Description
Technical field
The present embodiments relate to technical field of face recognition more particularly to a kind of face is blocked based on eigentransformation
Recognition methods and device.
Background technique
Recognition of face using more and more extensive, the requirement to recognition of face precision is higher and higher, especially in face
Divide the accuracy of identification in the case where being blocked even more important.
In the prior art, carrying out face usually using the face recognition algorithms based on convolutional neural networks is identification, still,
Face recognition algorithms based on convolutional neural networks are largely dependent upon the quality of data set, but face in actual scene
Block, in terms of have bigger complexity.It is also bigger by the artificial mark difficulty for blocking face picture, so mostly
The existing face recognition technologies of number are for by face picture is blocked, to there is a problem of that accuracy of identification declines serious.Existing raising face
Block the algorithm of robustness, the method for mostly using Multi net voting structure that plurality of human faces region is respectively trained greatly, by face different zones
Fusion Features, but there are computing resources to consume greatly for these methods, and the calculating time is longer, and the pretreatment of face picture is more complicated to be lacked
Point.
Summary of the invention
A kind of overcome the above problem the purpose of the embodiment of the present invention is that providing or at least be partially solved the above problem
Face identification method and device are blocked based on eigentransformation.
In order to solve the above-mentioned technical problem, on the one hand, the embodiment of the present invention provides a kind of being blocked based on eigentransformation
Face identification method, comprising:
Target facial image is input to preset convolutional neural networks model, most from the convolutional neural networks model
The latter convolutional layer obtains the characteristic pattern of the target facial image；
Calculate point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the target facial image
Feature for identification.
On the other hand, the embodiment of the present invention, which provides, a kind of is blocked face identification device based on eigentransformation, comprising:
Characteristic pattern extraction module, for target facial image to be input to preset convolutional neural networks model, from described
The last one convolutional layer of convolutional neural networks model obtains the characteristic pattern of the target facial image；
Computing module, for calculating point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Feature obtains module, defeated for described point of plain product to be input to the characteristic layer of the convolutional neural networks model
The feature for identification of the target facial image out.
In another aspect, the embodiment of the present invention provides a kind of electronic equipment, comprising:
Memory and processor, the processor and the memory complete mutual communication by bus；It is described to deposit
Reservoir is stored with the program instruction that can be executed by the processor, and it is above-mentioned that the processor calls described program instruction to be able to carry out
Method.
Another aspect, the embodiment of the present invention provide a kind of non-transient computer readable storage medium, are stored thereon with calculating
Machine program realizes above-mentioned method when the computer program is executed by processor.
It is provided in an embodiment of the present invention that face identification method and device are blocked based on eigentransformation, to being blocked face
Image using addition feature exposure mask eigentransformation mode, give up the common characteristic response vulnerable to occlusion area of face, engineering
Realization is easier, and the calculating time is shorter, and network structure is simpler, and is improved by the accuracy of identification for blocking face.
Detailed description of the invention
Fig. 1 is blocked face identification method schematic diagram based on eigentransformation to be provided in an embodiment of the present invention；
Fig. 2 is blocked face identification device schematic diagram based on eigentransformation to be provided in an embodiment of the present invention；
Fig. 3 is the structural schematic diagram of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
In order to keep the purposes, technical schemes and advantages of the embodiment of the present invention clearer, implement below in conjunction with the present invention
Attached drawing in example, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment
It is a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiment of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Fig. 1 is blocked face identification method schematic diagram based on eigentransformation to be provided in an embodiment of the present invention, such as Fig. 1 institute
Show, the embodiment of the present invention provide it is a kind of face identification method is blocked based on eigentransformation, executing subject is based on feature
Face identification device is blocked in transformation, which comprises
Step S101, target facial image is input to preset convolutional neural networks model, from the convolutional Neural net
The last one convolutional layer of network model obtains the characteristic pattern of the target facial image；
Step S102, point plain product of the characteristic pattern with the feature exposure mask obtained in advance is calculated；
Step S103, described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the mesh
Mark the feature for identification of facial image.
Specifically, provided in an embodiment of the present invention that face identification method, including two are blocked based on eigentransformation
Stage: one, experimental stage, two, application stage.
In the experimental stage, feature exposure mask is obtained according to experiment.
In the application stage, firstly, target facial image is input to preset convolutional neural networks model, from convolutional Neural
The last one convolutional layer of network model obtains the characteristic pattern of the target facial image.
Before the characteristic pattern input feature vector layer by the target facial image, the characteristic pattern of the target facial image is first calculated
With point plain product of the feature exposure mask obtained in advance.This feature exposure mask is to obtain in advance in the experimental stage.
Finally, point plain product being calculated is input to convolutional neural networks model as transformed characteristic pattern
Characteristic layer, export target facial image feature for identification, for identifying the target facial image according to this feature.
It is provided in an embodiment of the present invention that face identification method is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
On the basis of the above embodiments, further, obtaining the feature exposure mask, specific step is as follows:
Multiple original sample images that target data is concentrated are obtained, and artificial screening is added to the original sample image
Image after gear, the original sample image after adding and manually blocking, which is used as, blocks sample image, an original sample image
Corresponding one blocks sample image；
According to a pair of of original sample image and sample image is blocked, obtains an initial characteristics exposure mask；
The average value for taking all initial characteristics exposure masks, as final feature exposure mask.
Specifically, provided in an embodiment of the present invention that face identification method, including two are blocked based on eigentransformation
Stage: one, experimental stage, two, application stage.
In the experimental stage, feature exposure mask is obtained according to experiment.
Obtaining feature exposure mask, specific step is as follows:
Firstly, the multiple original sample images for obtaining target data concentration will be original for each original sample image
Sample image manually blocks face picture addition according to face characteristic point alignment, and according to characteristic point, obtains to the original sample
This image is added the image after manually blocking, and the original sample image after adding and manually blocking is used as and blocks sample graph
Picture, one original sample image corresponding one blocks sample image.
Then, according to a pair of of original sample image and sample image is blocked, an initial characteristics exposure mask is obtained.
Finally, the average value of all initial characteristics exposure masks is taken, as final feature exposure mask.
It is provided in an embodiment of the present invention that face identification method is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
It is further, described according to a pair of of original sample image and blocking sample graph on the basis of the above various embodiments
Picture obtains an initial characteristics mask feature exposure mask, specifically includes:
For target original sample image, the target original sample image is input to the convolutional neural networks mould
Type obtains the characteristic pattern of the target original sample image from the last one convolutional layer of the convolutional neural networks model, makees
For fisrt feature figure；And the corresponding target occlusion sample image of the target original sample image is input to the convolutional Neural
Network model obtains the feature of the target occlusion sample image from the last one convolutional layer of the convolutional neural networks model
Figure, as second feature figure；
It is poor that the fisrt feature figure and the second feature figure make, and obtains matrix of differences；
Binary conversion treatment is carried out to the element in the matrix of differences, obtains initial characteristics exposure mask.
Specifically, specific step is as follows for acquisition feature exposure mask:
Firstly, the multiple original sample images for obtaining target data concentration will be original for each original sample image
Sample image manually blocks face picture addition according to face characteristic point alignment, and according to characteristic point, obtains to the original sample
This image is added the image after manually blocking, and the original sample image after adding and manually blocking is used as and blocks sample graph
Picture, one original sample image corresponding one blocks sample image.
Then, according to a pair of of original sample image and sample image is blocked, an initial characteristics exposure mask is obtained.
Finally, the average value of all initial characteristics exposure masks is taken, as final feature exposure mask.
Wherein, according to a pair of of original sample image and sample image is blocked, an initial characteristics mask feature exposure mask is obtained,
It specifically includes:
For target original sample image, unobstructed target original sample image is input to convolutional neural networks mould
Type obtains the characteristic pattern of the target original sample image from the last one convolutional layer of convolutional neural networks model, as first
Characteristic pattern.
And the corresponding target occlusion sample image of target original sample image is input to convolutional neural networks model, from
The last one convolutional layer of convolutional neural networks model obtains the characteristic pattern of the target occlusion sample image, as second feature
Figure.
Then, it is poor fisrt feature figure and second feature figure make, and obtains matrix of differences.In matrix of differences, absolute value
Big point is to be blocked the characteristic response point of area sensitive to face.
Finally, carrying out binary conversion treatment by selecting suitable preset threshold to the element in the matrix of differences, obtaining just
Beginning feature exposure mask.The mode of binary conversion treatment are as follows: when element absolute value is greater than preset threshold, corresponding position 0；Element absolute value
When less than preset threshold, corresponding position 1.
It is provided in an embodiment of the present invention that face identification method is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
On the basis of the above various embodiments, further, the target data set is LFW data set.
Specifically, LFW data set is the human face data collection under a kind of unlimited environment, is widely used in the instruction of recognition of face
Practice.Target data set uses LFW data set, is that the feature exposure mask of acquisition is more accurate.
It is provided in an embodiment of the present invention that face identification method is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
On the basis of the above various embodiments, further, the convolutional neural networks model is resent-18 network.
Specifically, resent-18 network is a kind of simple, efficient convolutional neural networks model.The embodiment of the present invention
Using resent-18 network as convolutional neural networks model, the precision of recognition of face is improved.
It is provided in an embodiment of the present invention that face identification method is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
Fig. 2 is blocked face identification device schematic diagram based on eigentransformation to be provided in an embodiment of the present invention, such as Fig. 2 institute
Show, the embodiment of the present invention provide it is a kind of face identification device is blocked based on eigentransformation, for executing any of the above-described implementation
Method described in example specifically includes characteristic pattern extraction module 201, computing module 202 and feature and obtains module 203, in which:
Characteristic pattern extraction module 201 is used to target facial image being input to preset convolutional neural networks model, from institute
The last one convolutional layer for stating convolutional neural networks model obtains the characteristic pattern of the target facial image；Computing module 202 is used
In point plain product for calculating the characteristic pattern with the feature exposure mask obtained in advance；Feature obtains module 203 and is used to that described element will to be divided
Product is input to the characteristic layer of the convolutional neural networks model, exports the feature for identification of the target facial image.
Specifically, the face identification device that blocked provided in an embodiment of the present invention based on eigentransformation carries out face knowledge
Other process, including two stages: one, experimental stage, two, application stage.
In the experimental stage, feature exposure mask is obtained according to experiment.
In the application stage, firstly, target facial image is input to preset convolution mind by characteristic pattern extraction module 201
Through network model, the characteristic pattern of the target facial image is obtained from the last one convolutional layer of convolutional neural networks model.
Before the characteristic pattern input feature vector layer by the target facial image, first passes through computing module 202 and calculate the target
Point plain product of the characteristic pattern of facial image and the feature exposure mask obtained in advance.This feature exposure mask is to obtain in advance in the experimental stage
's.
Finally, obtaining module 203 using point plain product being calculated as defeated as transformed characteristic pattern by feature
Enter to the characteristic layer of convolutional neural networks model, the feature for identification of target facial image is exported, for according to this feature
Identify the target facial image.
It is provided in an embodiment of the present invention that face identification device is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
On the basis of the above embodiments, further, described device further includes that feature exposure mask obtains module, for obtaining
The feature exposure mask.
Specifically, provided in an embodiment of the present invention that face identification method, including two are blocked based on eigentransformation
Stage: one, experimental stage, two, application stage.
In the experimental stage, module is obtained by feature exposure mask, feature exposure mask is obtained according to experiment.
Obtaining feature exposure mask, specific step is as follows:
Firstly, the multiple original sample images for obtaining target data concentration will be original for each original sample image
Sample image manually blocks face picture addition according to face characteristic point alignment, and according to characteristic point, obtains to the original sample
This image is added the image after manually blocking, and the original sample image after adding and manually blocking is used as and blocks sample graph
Picture, one original sample image corresponding one blocks sample image.
Then, according to a pair of of original sample image and sample image is blocked, an initial characteristics exposure mask is obtained.
Finally, the average value of all initial characteristics exposure masks is taken, as final feature exposure mask.
It is provided in an embodiment of the present invention that face identification device is blocked based on eigentransformation, to by the image for blocking face
Using the eigentransformation mode of addition feature exposure mask, give up the common characteristic response vulnerable to occlusion area of face, Project Realization is more
It is easy, the calculating time is shorter, and network structure is simpler, and improves by the accuracy of identification for blocking face.
Fig. 3 is the structural schematic diagram of electronic equipment provided in an embodiment of the present invention, as shown in figure 3, the equipment includes: place
Manage device 301, memory 302 and bus 303；
Wherein, processor 301 and memory 302 complete mutual communication by the bus 303；
Processor 301 is used to call the program instruction in memory 302, to execute provided by above-mentioned each method embodiment
Method, for example,
Target facial image is input to preset convolutional neural networks model, most from the convolutional neural networks model
The latter convolutional layer obtains the characteristic pattern of the target facial image；
Calculate point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the target facial image
Feature for identification.
The embodiment of the present invention provides a kind of computer program product, and the computer program product is non-transient including being stored in
Computer program on computer readable storage medium, the computer program include program instruction, when described program instructs quilt
When computer executes, computer is able to carry out method provided by above-mentioned each method embodiment, for example,
Target facial image is input to preset convolutional neural networks model, most from the convolutional neural networks model
The latter convolutional layer obtains the characteristic pattern of the target facial image；
Calculate point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the target facial image
Feature for identification.
The embodiment of the present invention provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage
Medium storing computer instruction, the computer instruction make the computer execute side provided by above-mentioned each method embodiment
Method, for example,
Target facial image is input to preset convolutional neural networks model, most from the convolutional neural networks model
The latter convolutional layer obtains the characteristic pattern of the target facial image；
Calculate point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the target facial image
Feature for identification.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
The various media that can store program code such as disk.
The embodiments such as device and equipment described above are only schematical, wherein described be used as separate part description
Unit may or may not be physically separated, component shown as a unit may or may not be
Physical unit, it can it is in one place, or may be distributed over multiple network units.It can be according to the actual needs
Some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying
In the case where creative labor, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features；
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (9)
1. a kind of blocked face identification method based on eigentransformation characterized by comprising
Target facial image is input to preset convolutional neural networks model, from last of the convolutional neural networks model
A convolutional layer obtains the characteristic pattern of the target facial image；
Calculate point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Described point of plain product is input to the characteristic layer of the convolutional neural networks model, exports the use of the target facial image
In knowledge another characteristic.
2. the method according to claim 1, wherein obtaining the feature exposure mask, specific step is as follows:
Obtain target data concentrate multiple original sample images, and to the original sample image be added manually block after
Image, for the original sample image after adding and manually blocking as sample image is blocked, an original sample image is corresponding
One blocks sample image；
According to a pair of of original sample image and sample image is blocked, obtains an initial characteristics exposure mask；
The average value for taking all initial characteristics exposure masks, as final feature exposure mask.
3. according to the method described in claim 2, it is characterized in that, described according to a pair of of original sample image and blocking sample graph
Picture obtains an initial characteristics mask feature exposure mask, specifically includes:
For target original sample image, the target original sample image is input to the convolutional neural networks model, from
The last one convolutional layer of the convolutional neural networks model obtains the characteristic pattern of the target original sample image, as first
Characteristic pattern；And the corresponding target occlusion sample image of the target original sample image is input to the convolutional neural networks mould
Type obtains the characteristic pattern of the target occlusion sample image from the last one convolutional layer of the convolutional neural networks model, makees
For second feature figure；
It is poor that the fisrt feature figure and the second feature figure make, and obtains matrix of differences；
Binary conversion treatment is carried out to the element in the matrix of differences, obtains initial characteristics exposure mask.
4. method according to claim 1 or claim 2, which is characterized in that the target data set is LFW data set.
5. any one of -3 the method according to claim 1, which is characterized in that the convolutional neural networks model is resent-
18 networks.
6. a kind of blocked face identification device based on eigentransformation characterized by comprising
Characteristic pattern extraction module, for target facial image to be input to preset convolutional neural networks model, from the convolution
The last one convolutional layer of neural network model obtains the characteristic pattern of the target facial image；
Computing module, for calculating point plain product of the characteristic pattern with the feature exposure mask obtained in advance；
Feature obtains module, for described point of plain product to be input to the characteristic layer of the convolutional neural networks model, exports institute
State the feature for identification of target facial image.
7. device according to claim 6, which is characterized in that it further include that feature exposure mask obtains module, it is described for obtaining
Feature exposure mask.
8. a kind of electronic equipment characterized by comprising
Memory and processor, the processor and the memory complete mutual communication by bus；The memory
It is stored with the program instruction that can be executed by the processor, the processor calls described program instruction to be able to carry out right such as and wants
Seek 1 to 5 any method.
9. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that when the calculating
When machine program is executed by processor, method as claimed in claim 1 to 5 is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910006884.4A CN109840477B (en) | 2019-01-04 | 2019-01-04 | Method and device for recognizing shielded face based on feature transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910006884.4A CN109840477B (en) | 2019-01-04 | 2019-01-04 | Method and device for recognizing shielded face based on feature transformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109840477A true CN109840477A (en) | 2019-06-04 |
| CN109840477B CN109840477B (en) | 2020-11-24 |
Family
ID=66883669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910006884.4A Active CN109840477B (en) | 2019-01-04 | 2019-01-04 | Method and device for recognizing shielded face based on feature transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109840477B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287930A (en) * | 2019-07-01 | 2019-09-27 | 厦门美图之家科技有限公司 | Wrinkle disaggregated model training method and device |
| CN110647993A (en) * | 2019-09-23 | 2020-01-03 | 南方科技大学 | Infrared sensor mask manufacturing method, device and system and storage medium |
| CN110909595A (en) * | 2019-10-12 | 2020-03-24 | 平安科技（深圳）有限公司 | Facial motion recognition model training method and facial motion recognition method |
| CN111414879A (en) * | 2020-03-26 | 2020-07-14 | 北京字节跳动网络技术有限公司 | Face shielding degree identification method and device, electronic equipment and readable storage medium |
| CN111461959A (en) * | 2020-02-17 | 2020-07-28 | 浙江大学 | Face emotion synthesis method and device |
| CN111476200A (en) * | 2020-04-27 | 2020-07-31 | 华东师范大学 | Face de-identification generation method based on generation of confrontation network |
| CN111680597A (en) * | 2020-05-29 | 2020-09-18 | 北京百度网讯科技有限公司 | Face recognition model processing method, device, equipment and storage medium |
| CN111898413A (en) * | 2020-06-16 | 2020-11-06 | 深圳市雄帝科技股份有限公司 | Face recognition method, face recognition device, electronic equipment and medium |
| CN111931862A (en) * | 2020-09-11 | 2020-11-13 | 杭州追猎科技有限公司 | Method and system for detecting illegal posted advertisements and electronic equipment |
| CN112101302A (en) * | 2020-11-05 | 2020-12-18 | 杭州追猎科技有限公司 | Illegal poster detection method and system and electronic equipment |
| CN112149605A (en) * | 2020-09-30 | 2020-12-29 | 济南博观智能科技有限公司 | Face recognition method, device, equipment and storage medium |
| WO2022134337A1 (en) * | 2020-12-21 | 2022-06-30 | 平安科技（深圳）有限公司 | Face occlusion detection method and system, device, and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105095856A (en) * | 2015-06-26 | 2015-11-25 | 上海交通大学 | Method for recognizing human face with shielding based on mask layer |
| CN106203331A (en) * | 2016-07-08 | 2016-12-07 | 苏州平江历史街区保护整治有限责任公司 | A kind of crowd density evaluation method based on convolutional neural networks |
| CN106372595A (en) * | 2016-08-31 | 2017-02-01 | 重庆大学 | Shielded face identification method and device |
| CN108038435A (en) * | 2017-12-04 | 2018-05-15 | 中山大学 | A kind of feature extraction and method for tracking target based on convolutional neural networks |
| CN108509915A (en) * | 2018-04-03 | 2018-09-07 | 百度在线网络技术（北京）有限公司 | The generation method and device of human face recognition model |
-
2019
- 2019-01-04 CN CN201910006884.4A patent/CN109840477B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105095856A (en) * | 2015-06-26 | 2015-11-25 | 上海交通大学 | Method for recognizing human face with shielding based on mask layer |
| CN106203331A (en) * | 2016-07-08 | 2016-12-07 | 苏州平江历史街区保护整治有限责任公司 | A kind of crowd density evaluation method based on convolutional neural networks |
| CN106372595A (en) * | 2016-08-31 | 2017-02-01 | 重庆大学 | Shielded face identification method and device |
| CN108038435A (en) * | 2017-12-04 | 2018-05-15 | 中山大学 | A kind of feature extraction and method for tracking target based on convolutional neural networks |
| CN108509915A (en) * | 2018-04-03 | 2018-09-07 | 百度在线网络技术（北京）有限公司 | The generation method and device of human face recognition model |
Non-Patent Citations (1)
| Title |
|---|
| WEITAO WAN 等: "OCCLUSION ROBUST FACE RECOGNITION BASED ON MASK LEARNING", 《ICIP2017》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110287930A (en) * | 2019-07-01 | 2019-09-27 | 厦门美图之家科技有限公司 | Wrinkle disaggregated model training method and device |
| CN110647993A (en) * | 2019-09-23 | 2020-01-03 | 南方科技大学 | Infrared sensor mask manufacturing method, device and system and storage medium |
| CN110909595A (en) * | 2019-10-12 | 2020-03-24 | 平安科技（深圳）有限公司 | Facial motion recognition model training method and facial motion recognition method |
| CN110909595B (en) * | 2019-10-12 | 2023-04-18 | 平安科技（深圳）有限公司 | Facial motion recognition model training method and facial motion recognition method |
| WO2021068325A1 (en) * | 2019-10-12 | 2021-04-15 | 平安科技（深圳）有限公司 | Facial action recognition model training method, facial action recognition method and apparatus, computer device, and storage medium |
| CN111461959A (en) * | 2020-02-17 | 2020-07-28 | 浙江大学 | Face emotion synthesis method and device |
| CN111461959B (en) * | 2020-02-17 | 2023-04-25 | 浙江大学 | Face emotion synthesis method and device |
| CN111414879A (en) * | 2020-03-26 | 2020-07-14 | 北京字节跳动网络技术有限公司 | Face shielding degree identification method and device, electronic equipment and readable storage medium |
| CN111476200A (en) * | 2020-04-27 | 2020-07-31 | 华东师范大学 | Face de-identification generation method based on generation of confrontation network |
| CN111476200B (en) * | 2020-04-27 | 2022-04-19 | 华东师范大学 | Face de-identification generation method based on generation of confrontation network |
| CN111680597A (en) * | 2020-05-29 | 2020-09-18 | 北京百度网讯科技有限公司 | Face recognition model processing method, device, equipment and storage medium |
| CN111680597B (en) * | 2020-05-29 | 2023-09-01 | 北京百度网讯科技有限公司 | Face recognition model processing method, device, equipment and storage medium |
| CN111898413A (en) * | 2020-06-16 | 2020-11-06 | 深圳市雄帝科技股份有限公司 | Face recognition method, face recognition device, electronic equipment and medium |
| CN111931862A (en) * | 2020-09-11 | 2020-11-13 | 杭州追猎科技有限公司 | Method and system for detecting illegal posted advertisements and electronic equipment |
| CN111931862B (en) * | 2020-09-11 | 2021-07-23 | 杭州追猎科技有限公司 | Method and system for detecting illegal posted advertisements and electronic equipment |
| CN112149605A (en) * | 2020-09-30 | 2020-12-29 | 济南博观智能科技有限公司 | Face recognition method, device, equipment and storage medium |
| CN112149605B (en) * | 2020-09-30 | 2023-04-18 | 济南博观智能科技有限公司 | Face recognition method, device, equipment and storage medium |
| CN112101302B (en) * | 2020-11-05 | 2021-04-27 | 杭州追猎科技有限公司 | Illegal poster detection method and system and electronic equipment |
| CN112101302A (en) * | 2020-11-05 | 2020-12-18 | 杭州追猎科技有限公司 | Illegal poster detection method and system and electronic equipment |
| WO2022134337A1 (en) * | 2020-12-21 | 2022-06-30 | 平安科技（深圳）有限公司 | Face occlusion detection method and system, device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109840477B (en) | 2020-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840477A (en) | Face identification method and device are blocked based on eigentransformation | |
| CN109416727B (en) | Method and device for removing glasses in face image | |
| CN109325954B (en) | Image segmentation method and device and electronic equipment | |
| JP7520888B2 (en) | Method, apparatus, and non-transitory computer-readable storage medium for recognizing fake faces | |
| CN108765278A (en) | A kind of image processing method, mobile terminal and computer readable storage medium | |
| CN109241903A (en) | Sample data cleaning method, device, computer equipment and storage medium | |
| CN111340077B (en) | Attention mechanism-based disparity map acquisition method and device | |
| CN110378305B (en) | Tea disease identification method, equipment, storage medium and device | |
| JP6731529B1 (en) | Single-pixel attack sample generation method, device, equipment and storage medium | |
| CN110889855A (en) | Certificate photo matting method and system based on end-to-end convolutional neural network | |
| CN110378348A (en) | Instance of video dividing method, equipment and computer readable storage medium | |
| CN106709565A (en) | Neural network optimization method and device | |
| CN109214337A (en) | A kind of Demographics' method, apparatus, equipment and computer readable storage medium | |
| CN109165593A (en) | Feature extraction and matching and template renewal for biological identification | |
| CN109840485B (en) | Micro-expression feature extraction method, device, equipment and readable storage medium | |
| CN110046622B (en) | Targeted attack sample generation method, device, equipment and storage medium | |
| WO2020211242A1 (en) | Behavior recognition-based method, apparatus and storage medium | |
| CN109871845A (en) | Certificate image extracting method and terminal device | |
| Chen et al. | Automated design of neural network architectures with reinforcement learning for detection of global manipulations | |
| CN110059677A (en) | Digital table recognition methods and equipment based on deep learning | |
| CN109859314A (en) | Three-dimensional rebuilding method, device, electronic equipment and storage medium | |
| CN110263809A (en) | Pond characteristic pattern processing method, object detection method, system, device and medium | |
| CN110298394A (en) | A kind of image-recognizing method and relevant apparatus | |
| CN114783021A (en) | Intelligent detection method, device, equipment and medium for wearing of mask | |
| CN110297919A (en) | A kind of data cleaning method, device, equipment and storage medium |
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 |