CN104615997B - A kind of face method for anti-counterfeit based on multiple-camera - Google Patents
A kind of face method for anti-counterfeit based on multiple-camera Download PDFInfo
- Publication number
- CN104615997B CN104615997B CN201510080965.0A CN201510080965A CN104615997B CN 104615997 B CN104615997 B CN 104615997B CN 201510080965 A CN201510080965 A CN 201510080965A CN 104615997 B CN104615997 B CN 104615997B
- Authority
- CN
- China
- Prior art keywords
- face
- camera
- symmetrical
- facial image
- visual angle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to field of face identification, particularly a kind of face method for anti-counterfeit based on multiple-camera.The present invention is imaged from different perspectives using configuration multi-cam, and is carried out the method compared at face characteristic visual angle to the image of different cameras shooting and realized that recognition of face is anti-fake, and hardware configuration is simple, without calibration, is easy to be integrated with existing face identification system；Also it can differentiate that progress is anti-fake, effectively improves anti-spurious accuracy by merging a variety of face characteristic similarities simultaneously.
Description
Technical field
The present invention relates to field of face identification, particularly a kind of face method for anti-counterfeit based on multiple-camera.
Background technology
Two-dimension human face identification technology has obtained a large amount of use in the applications such as attendance and gate inhibition at present.But two-dimentional people
Face identifying system faces some problems in practical applications, such as cheats face identification system using photo or video.In order to
Solve the problems, such as this, face identification system method for anti-counterfeit comes into being.
Common face identification system method for anti-counterfeit includes the use of additional hardware and uses two major class of software algorithm.Wherein make
Generally judge whether object to be identified is true people using additional near infrared sensor is increased with the method for additional hardware
Face, and using the method for software algorithm then by analyzing the image of object to be identified or the feature of video (such as detection blink, table
End of love or behavior etc.) come distinguish real human face and photo etc. forge face.These methods introduce entirely different additional
Equipment is so as to increase the spy that the complexity of system and cost either depend on single image or the image sequence being continuously shot
Sign even requires user to perform certain action according to instruction, thus causes using upper inconvenience, also affects anti-fake essence
Degree.
Invention content
It is an object of the invention to the face method for anti-counterfeit currently based on multiple-camera is overcome to rely on to increase hardware or dependence
The problem of user's execute instruction acts, provides a kind of base for the anti-counterfeit capability and ease for use that can further improve face identification system
In the face method for anti-counterfeit of multiple-camera, comprise the following steps：
(1) n platform video cameras are symmetrical arranged and shoot facial image, n is more than 2 natural numbers.
(2) face feature point location is carried out respectively to the facial image of different cameras shooting, obtains each face feature
The coordinate of point.
(3) arbitrary three facial feature points not on the same line are selected, wherein fisrt feature point coordinates is (x1,
), y1 second feature point coordinates is (x2,y2), third feature point coordinates is (x3,y3), according to the people of each image of equation below calculating
Face visual angle characteristic d βi:
β1=arctan (y2-y1,x2-x1)×180/π；
β2=arctan (y3-(y1+y2)/2,x3-(x1+x2)/2)×180/π；
dβi=| β2-β1|。
(4) the face visual angle characteristic of the facial image of symmetrical camera shooting is compared, obtains face visual angle characteristic
Difference
(5) by face visual angle characteristic difference Δ and predetermined threshold value TΔIt is compared, as Δ ＞ TΔWhen, then it is judged as very, otherwise
It is false.
Further, the facial feature points include nose, the corners of the mouth, pupil center, canthus.
Preferably, in step (3), three facial feature points of selection are respectively Liang Ge pupil center and nose.
Further, it further includes and true and false step is judged by face characteristic similarity S as follows：
(6) texture eigenvalue T is extracted from the facial image that each video camera is shoti∈Rm, i is camera number, and m is
Intrinsic dimensionality.
(7) it calculates the facial image of each shot by camera and is in the shot by camera of symmetrical seat in the plane with it
Facial image between human face similarity degree Sk, calculation formula is：
Wherein, Ti,TjThe texture eigenvalue of respectively two positions symmetrical video camera mutually, the value range of k are
1 toWhen the value of p is 1 or 2, P value are 1, Euclidean distance is represented, when P values are 2, represent L1 distances.
(8) each facial image similarity S being calculatedkAverage value, as this shooting average human face similarity degree
S, calculation formula are：WhereinLower rounding for n/2.
(9) such as S ＞ TS, then it is judged as vacation, such as Δ ＞ TΔAnd S≤TSWhen, then it is determined as real human face, wherein TS is default
Face characteristic similarity threshold.
Further, in step (6), texture eigenvalue is takes local binary patterns (LBP), gradient orientation histogram mould
Formula (HOG) or Gabor filter patterns obtain initial value, and by the initial value of acquisition by principal component analysis or linear discriminant point
Analysis carries out dimensionality reduction or conversion process obtains.
In other embodiments, as α (Δ-TΔ)+β(TS- S) ＞ T when, be determined as real human face, be otherwise determined as
Face is forged, wherein α, β are constant, and meet alpha+beta=1；The threshold value that T is pre-set.
Further, the predetermined angle in n platforms video camera between adjacent camera is 0-45 degree.
In some embodiments, the n platforms video camera is in same level height.
In other embodiment, when the quantity of the video camera is even number, relative to same axis two between video camera
Two is symmetrical and form symmetrical group, and the video cameras that difference is symmetrically organized are in differentiated levels.
Further, the angle between two video cameras in different symmetrical groups is identical or differs.
In conclusion by adopting the above-described technical solution, the beneficial effects of the invention are as follows：The present invention takes the photograph using configuration
As head images from different perspectives, and carries out the method compared at face characteristic visual angle to the image of different cameras shooting and realize face
Identify anti-fake, hardware configuration is simple, without calibration, is easy to be integrated with existing face identification system；It simultaneously can also be by melting
It closes a variety of face characteristic similarities and differentiates that progress is anti-fake, effectively improve anti-spurious accuracy.
Description of the drawings
Fig. 1 is face visual angle characteristic method for anti-counterfeit flow chart provided by the invention.
Fig. 2 is the Double anti-counterfeit method integrated use flow chart provided in the embodiment of the present invention 2.
Fig. 3 is human face similarity degree method for anti-counterfeit flow chart in the Double anti-counterfeit method provided in the embodiment of the present invention 2.
Specific embodiment
Below in conjunction with the accompanying drawings, the present invention is described in detail.
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, it is right
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.
Embodiment 1：Such as Fig. 1, it is an object of the invention to the face method for anti-counterfeit currently based on multiple-camera is overcome to rely on to increase
Stiffened part relies on the problem of user's execute instruction acts, and provides a kind of anti-counterfeit capability that can further improve face identification system
With the face method for anti-counterfeit based on multiple-camera of ease for use, when video camera is even number, by taking 2 video cameras as an example, 2 are taken the photograph
Camera is set (preferably between 0-45 degree, in the present embodiment between two video cameras in same level height according to predetermined angle
Predetermined angle be 30 degree), and shoot facial image (should symmetrically shoot)；And comprising as follows by face visual angle characteristic difference Δ into
The step of row true and false judgement：
S101：Face feature point location is carried out respectively to the facial image of different cameras shooting, both determines each face
Characteristic point, as nose, the corners of the mouth, pupil center, canthus coordinate position.
S102：Liang Ge pupil center and nose are selected as three facial feature points, wherein the first center coordinate of eye pupil is
(x1,y1), the second center coordinate of eye pupil is (x2,y2), nose coordinate is (x3,y3), it is special to calculate face visual angle according to equation below
Levy d βi:
Calculate the angle of Liang Ge pupil center's lines and trunnion axis：β1=arctan (y2-y1,x2-x1)×180/π；
Calculate nose and the line of interpupillary line central point and the angle of trunnion axis：
β2=arctan (y3-(y1+y2)/2,x3-(x1+x2)/2)×180/π；
Calculate face visual angle characteristic：dβi=| β2-β1|。
S103：By the face visual angle characteristic d β of the facial image of two video camera shootings1、dβ2It is compared, obtains face
Visual angle characteristic difference Δ=| d β1-dβ2|。
S104：Face visual angle characteristic difference Δ and predetermined threshold value T Δs are compared, as Δ ＞ TΔWhen, then it is judged as very, it is no
It is then false.
Embodiment 2：When video camera is odd number, by taking 5 video cameras as an example, it is with 1 distinctive points of embodiment：
5 video cameras are set (preferably between 0-45 degree, in the present embodiment in same level height according to predetermined angle
Predetermined angle between adjacent two video cameras is 10 degree), and shoot facial image (marked as No. 1 camera shooting of 5 video camera sequences
Machine, No. 2 video cameras, No. 3 video cameras, No. 4 video cameras, No. 5 video cameras, face answers face to be located at No. 3 intermediate camera shootings during shooting
Machine, at this point, No. 1 video camera and No. 5 video cameras are symmetrical, No. 2 video cameras and No. 4 video cameras are symmetrical).
S101：Face feature point location is carried out respectively to the facial image of different cameras shooting, both determines each face
Characteristic point, as nose, the corners of the mouth, pupil center, canthus coordinate position.
S102：Liang Ge pupil center and nose are selected as three facial feature points, wherein the first center coordinate of eye pupil is
(x1,y1), the second center coordinate of eye pupil is (x2,y2), nose coordinate is (x3,y3), it is special to calculate face visual angle according to equation below
Levy d βi:
Calculate the angle of Liang Ge pupil center's lines and trunnion axis：β1=arctan (y2-y1,x2-x1)×180/π；
Calculate nose and the line of interpupillary line central point and the angle of trunnion axis：
β2=arctan (y3-(y1+y2)/2,x3-(x1+x2)/2)×180/π；
Calculate face visual angle characteristic：dβi=| β2-β1|。
S103：By the face visual angle characteristic d β of the facial image of 5 video camera shootings1、dβ2、dβ3、dβ4、dβ5Middle symmetry machine
The face visual angle characteristic of position is compared, and show that face visual angle characteristic is poorIn the present embodiment,In the visual angle characteristic of No. 3 video cameras wherein without symmetrical seat in the plane is not calculated in.This reality
It applies and only has 5 video cameras in example, therefore mutually symmetrical video camera is No. 1 video camera and No. 5 video cameras, No. 2 video cameras and No. 4
Video camera (video camera should be relative to facial symmetry during shooting).
S104：By face visual angle characteristic difference Δ and predetermined threshold value TΔIt is compared, as Δ ＞ TΔWhen, then it is judged as very, it is no
It is then false.
Embodiment 3：When video camera is even number, by taking 4 video cameras as an example, relative to same axis between 4 video cameras
Symmetrical two-by-two and 2 symmetrical group of composition, 2 video cameras symmetrically organized are in differentiated levels.Two in different symmetrical groups
Angle between video camera is different, if No. 1 video camera in the present embodiment and No. 4 video cameras are one symmetrical group, between the two
Angle is 30 degree, and No. 2 video cameras and No. 3 video cameras are one symmetrical group, and angle between the two is 45 degree；Face should during shooting
Face two is to the axis of symmetry of video camera.It is with 1 distinctive points of embodiment：
S103：By the face visual angle characteristic d β of the facial image of 4 video camera shootings1、dβ2、dβ3、dβ4In symmetrical seat in the plane
Face visual angle characteristic be compared, show that face visual angle characteristic is poorIn the present embodiment,
S104：By face visual angle characteristic difference Δ and predetermined threshold value TΔIt is compared, as Δ ＞ TΔWhen, then it is judged as very, it is no
It is then false.
Embodiment 4：As shown in Figure 2 and Figure 3, in the present embodiment, except including each step (letter as described in example 1 above
Referred to as face visual angle characteristic calculates) outside, further include the step of face characteristic similarity S is calculated as below：
S201：Texture eigenvalue T is extracted in the facial image shot from each video camerai∈Rm, i is camera number, this
The texture eigenvalue of the facial image of the shooting of two video cameras is respectively T in embodiment1And T2M is characterized dimension, and R is real number
Collection.
S202：It calculates the facial image of each shot by camera and is in the video camera of symmetrical seat in the plane with it and clapped
Human face similarity degree S between the facial image taken the photographK, only two video cameras, therefore calculation formula is in the present embodiment：
Wherein, T1,T2The texture eigenvalue of respectively two positions symmetrical video camera mutually, because only that 2 are taken the photograph
Camera, thereforeWhen the value of p is 1 or 2, P value are 1, Euclidean distance is represented, when P values are 2, represent L1 distances.
S203：Each facial image similarity S being calculatedKAverage value, as this shooting human face similarity degree S,
In the present embodiment, because only that 2 video cameras, therefore human face similarity degree S can be immediately arrived at by S202.
S204：Such as S ＞ TS, then it is judged as vacation, is otherwise true, wherein TSTo preset face characteristic similarity threshold.
In the present embodiment, because the face visual angle existed simultaneously described in face characteristic similarity calculation and embodiment 1 is special
The step of levying and calculate, therefore further including comprehensive descision, both (i.e. as Δ ＞ T only when both of which is judged as trueΔAnd simultaneously S≤
TSWhen), current face is just judged for real human face, is otherwise judged to forging face.
Embodiment 5：When video camera is odd number, by taking 5 video cameras as an example, except each including as described in example 2 above
Step (referred to as face visual angle characteristic calculates) outside, further includes the step of face characteristic similarity S is calculated as below：
S201：The texture eigenvalue of the facial image of the shooting of 5 video cameras is respectively T in the present embodiment1、T2、T3、T4、
T5, m is characterized dimension, and R is set of real numbers.
S202：It calculates the facial image of each shot by camera and is in the video camera of symmetrical seat in the plane with it and clapped
Human face similarity degree S between the facial image taken the photographi, only 5 video cameras in the present embodiment, therefore mutually symmetrical video camera is 1
Number video camera and No. 3 video cameras, No. 2 video cameras and No. 4 video cameras, and positioned at No. 3 intermediate video cameras due to there is no symmetry machine
Device, therefore the image of its shooting is not involved in similarity calculation；Calculation formula is：
Wherein, Ti,TjThe texture eigenvalue of respectively two positions symmetrical video camera mutually, the value model of wherein k
Enclose for 1 toWherein：
When the value of p is 1 or 2, P value are 1, Euclidean distance is represented, when P values are 2, represent L1 distances.
S203：Each facial image similarity S being calculatedkAverage value, as this shooting human face similarity degree S,
In the present embodiment,
S204：Such as S ＞ TS, then it is judged as vacation, is otherwise true, wherein TS is default face characteristic similarity threshold.
In the present embodiment, because the face visual angle existed simultaneously described in face characteristic similarity calculation and embodiment 2 is special
The step of levying the step of calculating, therefore further including comprehensive descision, both (i.e. as Δ ＞ T only when both of which is judged as trueΔIt is and same
When S≤TSWhen), current face is just judged for real human face, is otherwise judged to forging face.
Embodiment 6：When video camera is even number, by taking 4 video cameras as an example, except each including as described in example 3 above
Step (referred to as face visual angle characteristic calculates) outside, further includes the step of face characteristic similarity S is calculated as below：
S201：The texture eigenvalue of the facial image of the shooting of 4 video cameras is respectively T in the present embodiment1、T2、T3、T4,
M is characterized dimension, and R is set of real numbers.
S202：It calculates the facial image of each shot by camera and is in the video camera of symmetrical seat in the plane with it and clapped
Human face similarity degree S between the facial image taken the photographi, only 4 video cameras in the present embodiment, therefore mutually symmetrical video camera is 1
Number video camera and No. 4 video cameras, No. 2 video cameras and No. 3 video cameras, therefore image of its shooting is not involved in similarity calculation；Meter
Calculating formula is：
Wherein, Ti,TjThe texture eigenvalue of respectively two positions symmetrical video camera mutually, the value model of wherein k
Enclose for 1 toWherein：
When the value of p is 1 or 2, P value are 1, Euclidean distance is represented, when P values are 2, represent L1 distances.
S203：Each facial image similarity S being calculatedkAverage value, as this shooting human face similarity degree S,
In the present embodiment,
S204：Such as S ＞ TS, then it is judged as vacation, is otherwise true, wherein TSTo preset face characteristic similarity threshold.
In the present embodiment, because the face visual angle existed simultaneously described in face characteristic similarity calculation and embodiment 3 is special
The step of levying the step of calculating, therefore further including comprehensive descision, both (i.e. as Δ ＞ T only when both of which is judged as trueΔIt is and same
When S≤TSWhen), current face is just judged for real human face, is otherwise judged to forging face.
Embodiment 7：As shown in Figure 1, Figure 2, Figure 3 shows, in the present embodiment, compared with embodiment 4,5 or 6, difference lies in institutes
The step of stating comprehensive descision be：As α (Δ-TΔ)+β(TS- S) ＞ T when, be determined as real human face, be otherwise determined as forge face,
Wherein α, β are constant, and meet alpha+beta=1；The threshold value that T is pre-set.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement made within refreshing and principle etc., should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of face method for anti-counterfeit based on multiple-camera, which is characterized in that comprise the following steps：
(1) n platform video cameras are symmetrical arranged and shoot facial image, n is more than 2 natural numbers；
(2) face feature point location is carried out respectively to the facial image of different cameras shooting, obtains each facial feature points
Coordinate；
(3) arbitrary three facial feature points not on the same line are selected, wherein fisrt feature point coordinates is (x1,y1), second
Feature point coordinates is (x2,y2), third feature point coordinates is (x3,y3), the face visual angle that each image is calculated according to equation below is special
Levy d βi:
β1=arctan (y2-y1,x2-x1)×180/π；
β2=arctan (y3-(y1+y2)/2,x3-(x1+x2)/2)×180/π；
dβi=| β2-β1|；
(4) the face visual angle characteristic of the facial image of symmetrical camera shooting is compared, show that face visual angle characteristic is poor
(5) by face visual angle characteristic difference Δ and predetermined threshold value TΔIt is compared, as Δ ＞ TΔWhen, then it is judged as very, is otherwise false；
It further includes and true and false step is judged by face characteristic similarity S as follows：
(6) texture eigenvalue T is extracted from the facial image that each video camera is shoti∈Rm, i is camera number, and m is characterized
Dimension, R are set of real numbers；
(7) it calculates the facial image of each shot by camera and is in the people of the shot by camera of symmetrical seat in the plane with it
Human face similarity degree S between face imagek, calculation formula is：
Wherein, Ti,TjThe texture eigenvalue of respectively two positions symmetrical video camera mutually, the value range of k for 1 toThe value of p is 1 or 2；
(8) each facial image similarity S being calculatedkAverage value, as this shooting average human face similarity degree S；
(9) such as S ＞ TS, then it is judged as vacation, such as Δ ＞ TΔAnd S≤TSWhen, then it is determined as real human face, wherein TSTo preset face
Characteristic similarity threshold value.
2. the face method for anti-counterfeit based on multiple-camera as described in claim 1, which is characterized in that the facial feature points packet
Include nose, the corners of the mouth, pupil center, canthus.
3. the face method for anti-counterfeit based on multiple-camera as claimed in claim 2, which is characterized in that in step (3), selection
Three facial feature points are respectively Liang Ge pupil center and nose.
4. the face method for anti-counterfeit based on multiple-camera as described in claim 1, which is characterized in that in step (6), texture is special
Value indicative is that local binary patterns (LBP), gradient orientation histogram pattern (HOG) or Gabor filter patterns is taken to obtain initial value,
And the initial value of acquisition is obtained by principal component analysis or linear discriminant analysis progress dimensionality reduction or conversion process.
5. the face method for anti-counterfeit based on multiple-camera as described in claim 1, which is characterized in that as α (Δ-TΔ)+β(TS-
S) during ＞ T, it is determined as real human face, is otherwise judged to forging face, wherein α, β are constant, and meet alpha+beta=1；T is set in advance
The threshold value put.
6. the face method for anti-counterfeit based on multiple-camera as described in claim 1, which is characterized in that symmetrically arranged n platforms are taken the photograph
There is predetermined angle, which is 0-45 degree in camera between adjacent camera.
7. the face method for anti-counterfeit based on multiple-camera as described in claim 1, which is characterized in that at the n platforms video camera
In same level height.
8. the face method for anti-counterfeit based on multiple-camera as described in claim 1, which is characterized in that the quantity of the video camera
Symmetrical two-by-two relative to same axis between video camera and form symmetrical group during for even number, the video cameras that difference is symmetrically organized are in
Differentiated levels；
The angle between two video cameras in different symmetrical groups is identical or differs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510080965.0A CN104615997B (en) | 2015-02-15 | 2015-02-15 | A kind of face method for anti-counterfeit based on multiple-camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510080965.0A CN104615997B (en) | 2015-02-15 | 2015-02-15 | A kind of face method for anti-counterfeit based on multiple-camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104615997A CN104615997A (en) | 2015-05-13 |
| CN104615997B true CN104615997B (en) | 2018-06-19 |
Family
ID=53150434
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510080965.0A Active CN104615997B (en) | 2015-02-15 | 2015-02-15 | A kind of face method for anti-counterfeit based on multiple-camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104615997B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104881657B (en) * | 2015-06-08 | 2019-01-25 | 微梦创科网络科技（中国）有限公司 | Side face recognition methods, side face construction method and system |
| CN106355139B (en) * | 2016-08-22 | 2019-08-30 | 厦门中控智慧信息技术有限公司 | Face method for anti-counterfeit and device |
| CN106372629B (en) * | 2016-11-08 | 2020-02-07 | 汉王科技股份有限公司 | Living body detection method and device |
| CN108229326A (en) | 2017-03-16 | 2018-06-29 | 北京市商汤科技开发有限公司 | Face false-proof detection method and system, electronic equipment, program and medium |
| US10956714B2 (en) | 2018-05-18 | 2021-03-23 | Beijing Sensetime Technology Development Co., Ltd | Method and apparatus for detecting living body, electronic device, and storage medium |
| CN109359460B (en) * | 2018-11-20 | 2021-01-08 | 维沃移动通信有限公司 | Face recognition method and terminal equipment |
| CN112651319B (en) * | 2020-12-21 | 2023-12-05 | 科大讯飞股份有限公司 | Video detection method and device, electronic equipment and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110128385A1 (en) * | 2009-12-02 | 2011-06-02 | Honeywell International Inc. | Multi camera registration for high resolution target capture |
| CN101916496B (en) * | 2010-08-11 | 2013-10-02 | 无锡中星微电子有限公司 | System and method for detecting driving posture of driver |
| CN102254169B (en) * | 2011-08-23 | 2012-08-22 | 东北大学秦皇岛分校 | Multi-camera-based face recognition method and multi-camera-based face recognition system |
| CN102693418A (en) * | 2012-05-17 | 2012-09-26 | 上海中原电子技术工程有限公司 | Multi-pose face identification method and system |
| CN102779274B (en) * | 2012-07-19 | 2015-02-25 | 冠捷显示科技（厦门）有限公司 | Intelligent television face recognition method based on binocular camera |
-
2015
- 2015-02-15 CN CN201510080965.0A patent/CN104615997B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104615997A (en) | 2015-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104615997B (en) | A kind of face method for anti-counterfeit based on multiple-camera | |
| CN104364733B (en) | Watch position detecting device attentively, watch method for detecting position attentively and watch position detection program attentively | |
| JP4198951B2 (en) | Group attribute estimation method and group attribute estimation apparatus | |
| KR20220150868A (en) | Method of motion vector and feature vector based fake face detection and apparatus for the same | |
| CN109558764A (en) | Face identification method and device, computer equipment | |
| CN105447432B (en) | A kind of face method for anti-counterfeit based on local motion mode | |
| CN109840565A (en) | A kind of blink detection method based on eye contour feature point aspect ratio | |
| CN109359625A (en) | The method and system of customer identification is judged based on head and shoulder detection and face recognition technology | |
| JP6822482B2 (en) | Line-of-sight estimation device, line-of-sight estimation method, and program recording medium | |
| CN104573634A (en) | Three-dimensional face recognition method | |
| CN109670390A (en) | Living body face recognition method and system | |
| CN106355139B (en) | Face method for anti-counterfeit and device | |
| CN101071473A (en) | Feature point detector and its method | |
| CN108268814A (en) | A kind of face identification method and device based on the fusion of global and local feature Fuzzy | |
| KR101326691B1 (en) | Robust face recognition method through statistical learning of local features | |
| CN106650628B (en) | Fingertip detection method based on three-dimensional K curvature | |
| JP2021002840A (en) | Information processing system, information processor, information processing method and program | |
| Nurhudatiana et al. | On criminal identification in color skin images using skin marks (RPPVSM) and fusion with inferred vein patterns | |
| CN109426799A (en) | Information processing unit, storage medium and information processing method | |
| CN112633217A (en) | Human face recognition living body detection method for calculating sight direction based on three-dimensional eyeball model | |
| CN109146913A (en) | A kind of face tracking method and device | |
| CN109492584A (en) | A kind of recognition and tracking method and electronic equipment | |
| JP2008046691A (en) | Face image processor and program for computer | |
| CN108446639A (en) | Low-power consumption augmented reality equipment | |
| CN110929583A (en) | High-detection-precision face recognition method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |