CN109508702A - A kind of three-dimensional face biopsy method based on single image acquisition equipment - Google Patents
A kind of three-dimensional face biopsy method based on single image acquisition equipment Download PDFInfo
- Publication number
- CN109508702A CN109508702A CN201811638462.0A CN201811638462A CN109508702A CN 109508702 A CN109508702 A CN 109508702A CN 201811638462 A CN201811638462 A CN 201811638462A CN 109508702 A CN109508702 A CN 109508702A
- Authority
- CN
- China
- Prior art keywords
- characteristic point
- dimensional
- face
- user
- identified
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06V40/40—Spoof detection, e.g. liveness detection
- G06V40/45—Detection of the body part being alive
-
- 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
-
- 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)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention provides a kind of three-dimensional face biopsy methods based on single image acquisition equipment.Include: the recognition of face video flowing for acquiring user to be identified, judges whether there is face in initial baseline frame in recognition of face video flowing；If so, user to be identified is prompted to carry out simple yaw movement；Obtain multiple single-frame images of user to be identified recognition of face video flowing when carrying out compulsory exercise；Each characteristic point in each single-frame images is positioned, and obtains characteristic point coordinate；Solution is iterated to characteristic point coordinate, obtains characteristic point three-dimensional coordinate；All characteristic point three-dimensional coordinates of user to be identified are formed into space three-dimensional information, and In vivo detection is carried out to user to be identified using space three-dimensional information, obtain In vivo detection result.The present invention substantially increases the stability of In vivo detection in image recognition, reduces equipment cost, and testing conditions are realized simple.
Description
Technical field
The present invention relates to technical field of face recognition, acquire equipment based on single image more specifically to a kind of
Three-dimensional face biopsy method.
Background technique
In recent years, popularizing due to depth learning technology has carried out recognition of face using two-dimensional color or infrared picture
It is more mature, but consequent is the safety problem of recognition of face: it is directed to the face based on two dimensional technique such as gate inhibition, login
The means such as photo often can be used to identification carry out system deception in identifying system, user.Therefore, it is directed to the work of recognition of face
Physical examination is surveyed particularly important.
It is directed to In vivo detection at present, on 3-D technology, often carries out three-dimensional identification using double camera or polyphaser.
After binocular calibration, by characteristic point or matching algorithm, human face characteristic point or region are reconstructed.
But current is had the disadvantage in that using binocular technology progress three-dimensional reconstruction firstly, binocular (or more mesh) is for being
Structural stability of uniting is more demanding, if structure is because of Long-term Vibration or expands with heat and contract with cold repeatedly small variations occur, will have a direct impact on three
Tie up the output that face calculates；Secondly, dual camera systems, compared with one camera higher cost, structure, which designs, to be needed to pull open a certain distance
And it must assure that stringent synchronous acquisition on coefficient.These all constrain binocular or more mesh technologies in three-dimensional face identification skill significantly
Development in art.
Summary of the invention
In view of this, the present invention provide it is a kind of based on single image acquisition equipment three-dimensional face biopsy method to solve
Certainly the deficiencies in the prior art.
To solve the above problems, the present invention provides a kind of three-dimensional face In vivo detection side based on single image acquisition equipment
Method is applied to single image and acquires equipment, comprising:
The recognition of face video flowing for acquiring user to be identified judges in the initial baseline frame in the recognition of face video flowing
Whether face is had；
If having face in the initial baseline frame, the user to be identified is prompted to carry out compulsory exercise；
Obtain multiple single frames figures of the user to be identified recognition of face video flowing when carrying out the compulsory exercise
Picture；
Each characteristic point in each single-frame images is positioned, and obtains the corresponding characteristic point coordinate of the characteristic point；
Solution is iterated to the characteristic point coordinate, obtains the corresponding characteristic point three-dimensional coordinate of each characteristic point；
All characteristic point three-dimensional coordinates of the user to be identified are formed into space three-dimensional information, and utilize the sky
Between three-dimensional information In vivo detection is carried out to the user to be identified, obtain In vivo detection result.
Preferably, before described " the recognition of face video flowing for acquiring user to be identified ", further includes:
Image capture device is demarcated, the internal reference matrix of described image acquisition equipment is obtained.
Preferably, described " solution to be iterated to the characteristic point coordinate, obtains the corresponding characteristic point three of each characteristic point
Tie up coordinate " include:
Cost function is established, using the internal reference matrix, solution is iterated to the characteristic point coordinate, obtains each spy
The corresponding characteristic point three-dimensional coordinate of sign point；
The cost function are as follows:
Wherein, k is the internal reference matrix, and i is the number of the single-frame images, and j is characterized quantity a little；PiIt is i-th
Interframe projection matrix between single-frame images and the initial baseline frame, XjIt is sat for the corresponding characteristic point three-dimensional of j-th of characteristic point
Mark (xj,yj,zj, 1),For characteristic point coordinate of j-th of characteristic point in i-th single-frame images
Preferably, described " if having face in the initial baseline frame, to prompt the user to be identified to carry out regulation dynamic
Make " include:
If having face in the initial baseline frame, obtain between the user to be identified and described image acquisition equipment
Current relative distance；
Judge the current relative distance whether in default identification distance range；
If the current relative distance prompts the user to be identified to carry out regulation dynamic in default identification distance range
Make.
Preferably, the quantity of the characteristic point is at least 6；
The number of the single-frame images is at least 3.
Preferably, described " solution to be iterated to the characteristic point coordinate, obtains the corresponding characteristic point three of each characteristic point
Tie up coordinate " after, further includes:
By preset reference characteristic value, the scaling of the characteristic point three-dimensional coordinate is modified, after obtaining amendment
The characteristic point three-dimensional coordinate.
Preferably, the preset reference characteristic value is interpupillary distance.
In addition, to solve the above problems, the present invention also provides a kind of three-dimensional face based on single image acquisition equipment is living
Body detection device, comprising:
Acquisition module judges in the recognition of face video flowing for acquiring the recognition of face video flowing of user to be identified
Initial baseline frame in whether have face；
Cue module, if prompting the user to be identified to carry out regulation dynamic for having face in the initial baseline frame
Make；
Module is obtained, for obtaining the user to be identified recognition of face video flowing when carrying out the compulsory exercise
Multiple single-frame images；
Module is obtained, is also used to position each characteristic point in each single-frame images, and obtain the characteristic point pair
The characteristic point coordinate answered；
Computing module obtains the corresponding characteristic point of each characteristic point for being iterated solution to the characteristic point coordinate
Three-dimensional coordinate；
Detection module, for believing all characteristic point three-dimensional coordinate composition space three-dimensionals of the user to be identified
Breath, and In vivo detection is carried out to the user to be identified using the space three-dimensional information, obtain In vivo detection result.
In addition, to solve the above problems, the computer equipment includes storage the present invention also provides a kind of computer equipment
Device and processor；The memory is used to store the three-dimensional face In vivo detection program based on single image acquisition equipment, institute
The processor operation three-dimensional face In vivo detection program based on single image acquisition equipment is stated so that the computer equipment
Execute the three-dimensional face biopsy method as described above based on single image acquisition equipment.
In addition, to solve the above problems, the present invention also provides a kind of computer readable storage medium, it is described computer-readable
The three-dimensional face In vivo detection program based on single image acquisition equipment is stored on storage medium, it is described to be adopted based on single image
The three-dimensional face In vivo detection program of collection equipment is realized as described above based on single image acquisition equipment when being executed by processor
Three-dimensional face biopsy method.
A kind of three-dimensional face biopsy method based on single image acquisition equipment provided by the invention.Wherein, described
Method judges whether there is face in the initial baseline frame of video flowing by carrying out video acquisition to user to be identified, if tool
There is face, then after user carries out compulsory exercise, obtain the single-frame images of video flowing, and positions in the single-frame images of each frame
Multiple characteristic points, and then obtain the corresponding characteristic point coordinate in the picture of characteristic point and each feature is obtained by iterative solution
The characteristic point three-dimensional coordinate of point carries out user to be identified by the space three-dimensional information using three-dimensional feature point coordinate composition
In vivo detection.The present invention is realized through single image capture device, is carried out based on multiple characteristic points to user to be identified
Three-dimensional reconstruction obtains space three-dimensional information, and then carries out In vivo detection according to the space three-dimensional information, substantially increases image knowledge
The stability of not middle In vivo detection reduces equipment cost, and testing conditions are realized simply, are avoided in conventional identification techniques and are utilized
More image capture devices are identified that dual camera steady state accuracy requires high, multi-cam is at high cost, sets up camera to require
The cumbersome defect of the realization condition of distance.
Detailed description of the invention
Fig. 1 is what the three-dimensional face biopsy method example scheme for acquiring equipment the present invention is based on single image was related to
The structural schematic diagram of hardware running environment；
Fig. 2 is the process that the three-dimensional face biopsy method first embodiment of equipment is acquired the present invention is based on single image
Schematic diagram；
Fig. 3 is the principle that the three-dimensional face biopsy method second embodiment of equipment is acquired the present invention is based on single image
Schematic diagram；
Fig. 4 is the stream in the three-dimensional face biopsy method second embodiment for acquire equipment the present invention is based on single image
Journey schematic diagram；
Fig. 5 is the step of the present invention is based on the three-dimensional face biopsy method 3rd embodiments that single image acquires equipment
The flow diagram of S20 refinement；
Fig. 6 is to include the present invention is based on the three-dimensional face biopsy method 3rd embodiment of single image acquisition equipment
The flow diagram of step S70；
Fig. 7 is the functional module signal that the three-dimensional face living body detection device of equipment is acquired the present invention is based on single image
Figure.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, in which the same or similar labels are throughly indicated same or like
Element or element with the same or similar functions.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include one or more of the features.In the description of the present invention, the meaning of " plurality " is two or more,
Unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect
It connects, is also possible to be electrically connected；It can be directly connected, can also can be in two elements indirectly connected through an intermediary
The interaction relationship of the connection in portion or two elements.It for the ordinary skill in the art, can be according to specific feelings
Condition understands the concrete meaning of above-mentioned term in the present invention.
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
As shown in Figure 1, Fig. 1 is the structural schematic diagram of the hardware running environment for the terminal that the embodiment of the present invention is related to.
The PC that computer equipment of the embodiment of the present invention may each be is also possible to smart phone, tablet computer or portable
The packaged types terminal device such as computer.As shown in Figure 1, the computer equipment may include: processor 1001, such as CPU, net
Network interface 1004, user interface 1003, memory 1005, communication bus 1002.Wherein, communication bus 1002 is for realizing these
Connection communication between component.User interface 1003 may include display screen, input unit such as keyboard, remote controler, can be selected
Family interface 1003 can also include standard wireline interface and wireless interface.Network interface 1004 optionally may include standard
Wireline interface, wireless interface (such as WI-FI interface).Memory 1005 can be high speed RAM memory, be also possible to stable deposit
Reservoir, such as magnetic disk storage.Memory 1005 optionally can also be the storage device independently of aforementioned processor 1001.It can
Selection of land, computer equipment can also include RF (Radio Frequency, radio frequency) circuit, voicefrequency circuit, WiFi module etc..
In addition, computer equipment can also configure the other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor,
Details are not described herein.
It will be understood by those skilled in the art that computer equipment shown in Fig. 1 does not constitute the restriction to it, can wrap
It includes than illustrating more or fewer components, perhaps combines certain components or different component layouts.As shown in Figure 1, as one
It may include operating system, data-interface control program, network connection in the memory 1005 of kind computer readable storage medium
Program and the three-dimensional face In vivo detection program that equipment is acquired based on single image.
A kind of three-dimensional face biopsy method based on single image acquisition equipment provided by the invention.Wherein, described
Method is realized through single image capture device, is carried out three-dimensional reconstruction based on multiple characteristic points to user to be identified, is obtained
In vivo detection is carried out to space three-dimensional information, and then according to the space three-dimensional information, living body in image recognition is substantially increased and examines
The stability of survey, reduces equipment cost, and testing conditions are realized simple.
Embodiment 1:
Referring to Fig. 2, first embodiment of the invention provides a kind of three-dimensional face living body inspection based on single image acquisition equipment
Survey method, comprising:
Step S10 acquires the recognition of face video flowing of user to be identified, judges initial in the recognition of face video flowing
Whether there is face in reference frame；
Above-mentioned, the present embodiment is set applied to single image capture device, i.e. one camera by single Image Acquisition
The standby three-dimensional reconstruction realized for user to be identified, to further carry out the purpose of In vivo detection to user to be identified.
It is above-mentioned, recognition of face video flowing, for by the single collected video flowing of image capture device.
It is above-mentioned, judge that the key frame in video flowing can be obtained with the presence or absence of face in video flowing, to carrying out face in key frame
Portion's feature location, for example, being positioned for eyes, nose and mouth according to characteristics of image, if wherein including above-mentioned organ pair
The characteristics of image answered, and the distance between characteristics of image then determines the frame video within default face organ's distance range
There are faces in stream.
It is above-mentioned, initial baseline frame, as in video flowing, the first frame image that is identified.
Step S20 prompts the user to be identified to carry out compulsory exercise if having face in the initial baseline frame；
Movement above-mentioned, that the compulsory exercise, as prompt user are completed, for example, shaking the head, nodding, blinking, turning
Head moves up and down head etc. to make head (face) changed movement in the visual field of image capture device.
It is above-mentioned, it prompts, as by the way that the three-dimensional face biopsy method based on single image acquisition equipment can be achieved
Device issues the prompt information for carrying out showing compulsory exercise, such as prompt user to user to be identified according to the personage in screen
It influences, does identical movement.
In addition, can return to the step S10 if face is not present in the initial baseline frame.
Step S30, obtain the user to be identified when carrying out the compulsory exercise recognition of face video flowing it is more
A single-frame images；
Step S40 positions each characteristic point in each single-frame images, and obtains the corresponding feature of the characteristic point
Point coordinate；
Step S50 is iterated solution to the characteristic point coordinate, obtains the corresponding characteristic point three-dimensional of each characteristic point and sits
Mark；
Above-mentioned, single-frame images, for the image of the multiple single frames obtained in video flowing, quantity is more, then carries out to user
In vivo detection judges more accurate.
It is above-mentioned, it include the face of user to be identified in each single-frame images.Wherein, characteristic point, to be preset convenient for knowledge
Other user's uniqueness, either convenient for the characteristic portion or feature locations of acquisition measurement data, for example, the corners of the mouth, canthus, nose
Point, eyebrow etc..And then selected characteristic point is tracked, to carry out further In vivo detection.
Above-mentioned, using initial detecting frame as initial baseline frame, user acts move according to the rules, it is assumed that initial baseline frame
On, the three-dimensional coordinate (being reference with camera coordinates system) of some characteristic point is W, on initial baseline frame, the image of this feature point
Coordinate is m, then in subsequent any one frame, due to the movement of face, the evolution of this feature point on the image to m '.Transformation
Reference frame, is considered as camera motion and face is motionless, and the position of characteristic point is caused to be changed.And camera motion can
It is described by spin matrix R and translation matrix T.It then can be according to above-mentioned variation, to three based on characteristic point of user to be identified
Dimension is rebuild.That is, causing characteristic point spatially to change according to the movement of user, it can be considered image capture device in space
On change, and the characteristic point of user's face is spatially constant, so that iterative calculation solves to the characteristic point coordinate of single frames
Obtain characteristic point three-dimensional coordinate.
All characteristic point three-dimensional coordinates of the user to be identified are formed space three-dimensional information, and benefit by step S60
In vivo detection is carried out to the user to be identified with the space three-dimensional information, obtains In vivo detection result.
It is above-mentioned, In vivo detection is carried out using the three-dimensional information of face, can effectively avoid because of the frauds bring such as photo
Risk.In vivo detection is carried out to the user to be identified using the space three-dimensional information, obtain In vivo detection as a result, for example,
The processing means such as plane fitting can be carried out to the corresponding characteristic point three-dimensional coordinate of characteristic point and whether carry out identification feature point from one
A plane, or according to the information such as canthus, the corners of the mouth, nose and color image, face is modeled, verify the distribution of its face
Reasonability etc. carries out In vivo detection, to further reach the confirmation purpose for living body.
A kind of three-dimensional face biopsy method based on single image acquisition equipment provided in this embodiment.By treating
Identify that user carries out video acquisition, and judge whether there is face in the initial baseline frame of video flowing, if having face, with
After family carries out compulsory exercise, the single-frame images of video flowing is obtained, and positions multiple characteristic points in the single-frame images of each frame, into
And the corresponding characteristic point coordinate in the picture of characteristic point is obtained, by iterative solution, obtain the characteristic point three of each characteristic point
Coordinate is tieed up, by the space three-dimensional information using three-dimensional feature point coordinate composition, In vivo detection is carried out to user to be identified.This reality
Applying example realizes through single image capture device, carries out three-dimensional reconstruction based on multiple characteristic points to user to be identified, obtains
In vivo detection is carried out to space three-dimensional information, and then according to the space three-dimensional information, living body in image recognition is substantially increased and examines
The stability of survey reduces equipment cost, and testing conditions are realized simply, avoids and utilizes more Image Acquisition in conventional identification techniques
Equipment is identified that dual camera steady state accuracy requires the realization that high, multi-cam is at high cost, sets up camera requirement distance
The cumbersome defect of condition.
Embodiment 2:
Referring to Fig. 3-4, second embodiment of the invention provides a kind of three-dimensional face living body based on single image acquisition equipment
Detection method, is based on above-mentioned first embodiment shown in Fig. 2, and the step S10 " acquires the recognition of face view of user to be identified
Frequency flows " before, further includes:
Step S60 demarcates described image acquisition equipment, obtains the internal reference matrix of described image acquisition equipment.
Above-mentioned, image capture device can be camera, or high-definition camera, in the present embodiment, individually to set
It is standby.Single camera is demarcated, as to the pretreatment of equipment, obtains the internal reference matrix K of camera, which is usually using
It is completed before camera default.
The step S50 " is iterated solution to the characteristic point coordinate, obtains the corresponding characteristic point of each characteristic point
Three-dimensional coordinate " includes:
Step S51, establishes cost function, using the internal reference matrix, is iterated solution to the characteristic point coordinate, obtains
To the corresponding characteristic point three-dimensional coordinate of each characteristic point；
The cost function are as follows:
Wherein, k is the internal reference matrix, and i is the number of the single-frame images, and j is characterized quantity a little；PiIt is i-th
Interframe projection matrix between single-frame images and the initial baseline frame, XjIt is sat for the corresponding characteristic point three-dimensional of j-th of characteristic point
Mark (xj,yj,zj, 1),For characteristic point coordinate of j-th of characteristic point in i-th single-frame images
Using initial detecting frame as benchmark frame, it is assumed that on reference frame, the three-dimensional coordinate of some characteristic point (is with camera coordinates system
With reference to) it is W (x, y, z), on reference frame, the image coordinate of this feature point is m (u, v), then in subsequent any one frame, due to
The movement of face, the evolution of this feature point on the image to m ' (u ', v ').Reference frame is converted, can regard camera as
Movement and face is motionless, cause the position of characteristic point to be changed.And camera motion can be by spin matrix R and translation matrix T
(tx,ty,tz) be described, as shown in Figure 3:
Also, it is based on principle, the image coordinate and three-dimensional coordinate of the characteristic point on reference frame and arbitrary frame meet such as ShiShimonoseki
System:
m′norm=(u ', v ', 1)；
Wherein:P=[T× TK-TT×RK-1, T], it is interframe projection matrix.
Wherein:Referred to as T's is oblique at matrix.
Unknown number is characterized R, T matrix of three-dimensional coordinate W and interframe a little.By above-mentioned equation, below in following steps
Cost function solves the unknown number.Acquire the value that R, T and W make the E value minimum of following formula that can solve key point three-dimensional coordinate W:
Wherein, i is image number, and j is that the face chosen closes
The number of key point.PiFor i-th interframe projection matrix between image and benchmark image, XjFor the three-dimensional coordinate of j-th of key point
(xj,yj,zj, 1),For coordinate of j-th of key point on i-th imageAbove formula is Solving Nonlinear Equation
Process can more fast and accurately be calculated result such as L-M iterative method, Newton iteration method etc., be improved most using iterative method
The accuracy of result is obtained eventually.By iterative solution, the corresponding characteristic point three-dimensional coordinate of each characteristic point of face can be obtained.
Embodiment 3:
Referring to Fig. 5-6, third embodiment of the invention provides a kind of three-dimensional face living body based on single image acquisition equipment
Detection method, be based on above-mentioned second embodiment shown in Fig. 4, the step S20, " if having face in the initial baseline frame,
The user to be identified is prompted to carry out compulsory exercise " include:
Step S21 obtains the user to be identified and sets with described image acquisition if having face in the initial baseline frame
Current relative distance between standby；
Whether step S22 judges the current relative distance in default identification distance range；
Step S23, if the current relative distance in default identification distance range, prompt the user to be identified into
Row compulsory exercise.
It is above-mentioned, before carrying out In vivo detection, to guarantee that the position for obtaining facial image in video flowing convenient for quickly identification, needs
User to be identified and image capture device is wanted to be maintained within default identification distance range, hypertelorism then will lead to face figure
As the problems such as less than normal, image obscures, feature is unintelligible, and hypotelorism, then it may result in facial image and show incomplete, display
The problems such as light difference etc., so to preset a default identification distance range, to guarantee user within that range, Huo Zhegai
The distance between user and image capture device are in the range.
Specifically, the judgement of distance or range, can by image recognition, i.e., by the user position in video flowing into
Row calibration, so that the calibration data are compared with preset data, to obtain a relative distance, retell relative distance in advance
If identification distance range is compared, to can determine whether the user is in default identification distance range.
In addition, can also carry out judging the position of user by other means, for example, passing through infrared sensing
Device triggers infrared sensor, is generated so that system is received by user's trigger sensor after user reaches in specified region
Instruction, according to instruction determine user it is in place, further video flowing acquisition can be carried out.In the present embodiment, by into
Before row is to user's In vivo detection, judge first whether user is in default identification distance range, so that it is guaranteed that user
Location is in the range of being accurately identified, to improve the accuracy and identification effect for user's identification
Rate.
In addition, if the current relative distance not in default identification distance range, prompt the user to be identified into
Line position sets adjustment, and returns to the step S22.
The quantity of the characteristic point is at least 6；The number of the single-frame images is at least 3.
It is above-mentioned, it is contemplated that above-mentioned cost function, equation number need to be greater than the quantity of unknown number, can just be iterated solution, then
The number that the quantity of the characteristic point is at least 6 single-frame images is at least 3.
In addition, in order to more accurately be calculated, it is proposed that characteristic point is greater than 12, and the number of single-frame images is greater than 20.
The step S50 " is iterated solution to the characteristic point coordinate, obtains the corresponding characteristic point of each characteristic point
After three-dimensional coordinate ", further includes:
Step S70 is modified the scaling of the characteristic point three-dimensional coordinate, is obtained by preset reference characteristic value
To the revised characteristic point three-dimensional coordinate.
The preset reference characteristic value is interpupillary distance.
In addition it is also possible to represent the value of facial dimension feature for other, for example, interpupillary distance corners of the mouth distance, bridge of the nose length,
Face's most width between, left and right distalmost end canthus with a distance from etc..
After being iterated solution, the characteristic point three-dimensional coordinate of user's face is obtained.And the three-dimensional of the characteristic point obtained is sat
W is marked, usually there are certain scaling relationships with real world coordinates system.
For example, fixed interpupillary distance (as assumed to be fixed value between two canthus) can be used in preset reference characteristic value, to calculating
Three-dimensional coordinate W out is zoomed in and out again, is allowed to close to true three-dimensional world coordinate.
Multiple characteristic point three-dimensional coordinates of user to be identified have spatial relation as a whole from each other,
But its scaling, with user's actual face ratio, it is understood that there may be certain difference in the present embodiment, passes through preset reference feature
Value is modified the characteristic point three-dimensional coordinate of user to be identified obtained, between adjustment scaling this feature point three-dimensional coordinate
Spatial relation, to be allowed to be more nearly the three-dimensional coordinate of user's actual spatial relation and ratio, to obtain
More accurate three-dimensional reconstruction result improves the accuracy of identification.
In addition, the present invention also provides a kind of three-dimensional face In vivo detection dresses based on single image acquisition equipment with reference to Fig. 7
It sets, comprising:
Acquisition module 10 judges the recognition of face video flowing for acquiring the recognition of face video flowing of user to be identified
In initial baseline frame in whether have face；
Cue module 20, if prompting the user to be identified to provide for having face in the initial baseline frame
Movement；
Module 30 is obtained, for obtaining the user to be identified recognition of face video when carrying out the compulsory exercise
Multiple single-frame images of stream；
Module 30 is obtained, is also used to position each characteristic point in each single-frame images, and obtain the characteristic point
Corresponding characteristic point coordinate；
Computing module 40 obtains the corresponding feature of each characteristic point for being iterated solution to the characteristic point coordinate
Point three-dimensional coordinate；
Detection module 50, for believing all characteristic point three-dimensional coordinate composition space three-dimensionals of the user to be identified
Breath, and In vivo detection is carried out to the user to be identified using the space three-dimensional information, obtain In vivo detection result.
In addition, the computer equipment includes memory and processor the present invention also provides a kind of computer equipment；Institute
Memory is stated for store the three-dimensional face In vivo detection program based on single image acquisition equipment, described in the processor is run
Three-dimensional face In vivo detection program based on single image acquisition equipment is so that the computer equipment executes base as described above
In the three-dimensional face biopsy method of single image acquisition equipment.
In addition, being stored on the computer readable storage medium the present invention also provides a kind of computer readable storage medium
There are the three-dimensional face In vivo detection program based on single image acquisition equipment, the three-dimensional people based on single image acquisition equipment
Face In vivo detection program realizes the three-dimensional face living body as described above based on single image acquisition equipment when being executed by processor
Detection method.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.Pass through above embodiment party
The description of formula, it is required general that those skilled in the art can be understood that above-described embodiment method can add by software
The mode of hardware platform is realized, naturally it is also possible to which by hardware, but in many cases, the former is more preferably embodiment.It is based on
Such understanding, substantially the part that contributes to existing technology can be with software product in other words for technical solution of the present invention
Form embody, which is stored in a storage medium (such as ROM/RAM, magnetic disk, light as described above
Disk) in, including some instructions use is so that a terminal device (can be mobile phone, computer, server or the network equipment
Deng) execute method described in each embodiment of the present invention.The above is only a preferred embodiment of the present invention, is not intended to limit this hair
Bright the scope of the patents, it is all using equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content, or directly
It connects or is used in other related technical areas indirectly, be included within the scope of the present invention.
Claims (10)
1. a kind of three-dimensional face biopsy method based on single image acquisition equipment characterized by comprising
The recognition of face video flowing for acquiring user to be identified, judge in the initial baseline frame in the recognition of face video flowing whether
There is face；
If having face in the initial baseline frame, the user to be identified is prompted to carry out compulsory exercise；
Obtain multiple single-frame images of the user to be identified recognition of face video flowing when carrying out the compulsory exercise；
Each characteristic point in each single-frame images is positioned, and obtains the corresponding characteristic point coordinate of the characteristic point；
Solution is iterated to the characteristic point coordinate, obtains the corresponding characteristic point three-dimensional coordinate of each characteristic point；
All characteristic point three-dimensional coordinates of the user to be identified are formed into space three-dimensional information, and utilize the space three
It ties up information and In vivo detection is carried out to the user to be identified, obtain In vivo detection result.
2. the three-dimensional face biopsy method as described in claim 1 based on single image acquisition equipment, which is characterized in that institute
Before stating " the recognition of face video flowing for acquiring user to be identified ", further includes:
Described image acquisition equipment is demarcated, the internal reference matrix of described image acquisition equipment is obtained.
3. the three-dimensional face biopsy method as claimed in claim 2 based on single image acquisition equipment, which is characterized in that institute
Stating " being iterated solution to the characteristic point coordinate, obtain the corresponding characteristic point three-dimensional coordinate of each characteristic point " includes:
Cost function is established, using the internal reference matrix, solution is iterated to the characteristic point coordinate, obtains each characteristic point
Corresponding characteristic point three-dimensional coordinate；
The cost function are as follows:
Wherein, k is the internal reference matrix, and i is the number of the single-frame images, and j is characterized quantity a little；PiFor i-th single frames figure
Picture and the interframe projection matrix between the initial baseline frame, XjFor the corresponding characteristic point three-dimensional coordinate (x of j-th of characteristic pointj,
yj,zj, 1),For characteristic point coordinate of j-th of characteristic point in i-th single-frame images
4. the three-dimensional face biopsy method as claimed in claim 3 based on single image acquisition equipment, which is characterized in that institute
Stating " if having face in the initial baseline frame, the user to be identified being prompted to carry out compulsory exercise " includes:
If having face in the initial baseline frame, obtain current between the user to be identified and described image acquisition equipment
Relative distance；
Judge the current relative distance whether in default identification distance range；
If the current relative distance prompts the user to be identified to carry out compulsory exercise in default identification distance range.
5. the three-dimensional face biopsy method as described in claim 1 based on single image acquisition equipment, which is characterized in that institute
The quantity for stating characteristic point is at least 6；
The number of the single-frame images is at least 3.
6. the three-dimensional face biopsy method as described in claim 1 based on single image acquisition equipment, which is characterized in that institute
After stating " being iterated solution to the characteristic point coordinate, obtain the corresponding characteristic point three-dimensional coordinate of each characteristic point ", also wrap
It includes:
By preset reference characteristic value, the scaling of the characteristic point three-dimensional coordinate is modified, obtains revised institute
State characteristic point three-dimensional coordinate.
7. the three-dimensional face biopsy method as claimed in claim 6 based on single image acquisition equipment, which is characterized in that institute
Stating preset reference characteristic value is interpupillary distance.
8. a kind of three-dimensional face living body detection device based on single image acquisition equipment characterized by comprising
Acquisition module judges first in the recognition of face video flowing for acquiring the recognition of face video flowing of user to be identified
Whether there is face in beginning reference frame；
Cue module, if prompting the user to be identified to carry out compulsory exercise for having face in the initial baseline frame；
Obtain module, for obtain the user to be identified when carrying out the compulsory exercise recognition of face video flowing it is more
A single-frame images；
The acquisition module is also used to position each characteristic point in each single-frame images, and obtains the characteristic point pair
The characteristic point coordinate answered；
It is three-dimensional to obtain the corresponding characteristic point of each characteristic point for being iterated solution to the characteristic point coordinate for computing module
Coordinate；
Detection module, for all characteristic point three-dimensional coordinates of the user to be identified to be formed space three-dimensional information, and
In vivo detection is carried out to the user to be identified using the space three-dimensional information, obtains In vivo detection result.
9. a kind of computer equipment, which is characterized in that the computer equipment includes memory and processor；The memory
For storing the three-dimensional face In vivo detection program based on single image acquisition equipment, the processor operation is described based on single
The three-dimensional face In vivo detection program of image capture device is so that the computer equipment is executed such as any one of claim 1-7
The three-dimensional face biopsy method based on single image acquisition equipment.
10. a kind of computer readable storage medium, which is characterized in that be stored on the computer readable storage medium based on single
The three-dimensional face In vivo detection program of one image capture device, the three-dimensional face living body inspection based on single image acquisition equipment
The three-dimensional people based on single image acquisition equipment as described in any one of claim 1-7 is realized when ranging sequence is executed by processor
Face biopsy method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811638462.0A CN109508702A (en) | 2018-12-29 | 2018-12-29 | A kind of three-dimensional face biopsy method based on single image acquisition equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811638462.0A CN109508702A (en) | 2018-12-29 | 2018-12-29 | A kind of three-dimensional face biopsy method based on single image acquisition equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109508702A true CN109508702A (en) | 2019-03-22 |
Family
ID=65757059
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811638462.0A Pending CN109508702A (en) | 2018-12-29 | 2018-12-29 | A kind of three-dimensional face biopsy method based on single image acquisition equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109508702A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110164013A (en) * | 2019-05-31 | 2019-08-23 | 广州软盈科技有限公司 | A kind of retrospective caller management method |
| CN110222573A (en) * | 2019-05-07 | 2019-09-10 | 平安科技（深圳）有限公司 | Face identification method, device, computer equipment and storage medium |
| CN110544335A (en) * | 2019-08-30 | 2019-12-06 | 北京市商汤科技开发有限公司 | Object recognition system and method, electronic device, and storage medium |
| CN111339958A (en) * | 2020-02-28 | 2020-06-26 | 山东笛卡尔智能科技有限公司 | Monocular vision-based face in-vivo detection method and system |
| CN111783599A (en) * | 2020-06-24 | 2020-10-16 | 东风汽车有限公司 | Face living body identification method and electronic equipment |
| CN111860057A (en) * | 2019-04-29 | 2020-10-30 | 北京眼神智能科技有限公司 | Face image blurring and living body detection method and device, storage medium and equipment |
| CN112016437A (en) * | 2020-08-26 | 2020-12-01 | 中国科学院重庆绿色智能技术研究院 | Living body detection method based on face video key frame |
| CN112418036A (en) * | 2020-11-12 | 2021-02-26 | 广州市保伦电子有限公司 | Video conference order analysis method and processing terminal |
| CN112487921A (en) * | 2020-11-25 | 2021-03-12 | 奥比中光科技集团股份有限公司 | Face image preprocessing method and system for living body detection |
| CN112711968A (en) * | 2019-10-24 | 2021-04-27 | 浙江舜宇智能光学技术有限公司 | Face living body detection method and system |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101839692A (en) * | 2010-05-27 | 2010-09-22 | 西安交通大学 | Method for measuring three-dimensional position and stance of object with single camera |
| CN101908140A (en) * | 2010-07-29 | 2010-12-08 | 中山大学 | Biopsy method for use in human face identification |
| CN102594964A (en) * | 2012-02-13 | 2012-07-18 | 广东步步高电子工业有限公司 | Intelligent contact adding method using mobile terminal |
| CN103824278A (en) * | 2013-12-10 | 2014-05-28 | 清华大学 | Monitoring camera calibration method and system |
| CN105205438A (en) * | 2014-09-05 | 2015-12-30 | 北京七鑫易维信息技术有限公司 | Method of using infrared eyeball to track and control distance of eyes and screen and system thereof |
| CN105260726A (en) * | 2015-11-11 | 2016-01-20 | 杭州海量信息技术有限公司 | Interactive video in vivo detection method based on face attitude control and system thereof |
| CN105740781A (en) * | 2016-01-25 | 2016-07-06 | 北京天诚盛业科技有限公司 | Three-dimensional human face in-vivo detection method and device |
| CN105868677A (en) * | 2015-01-19 | 2016-08-17 | 阿里巴巴集团控股有限公司 | Live human face detection method and device |
| CN106529414A (en) * | 2016-10-14 | 2017-03-22 | 国政通科技股份有限公司 | Method for realizing result authentication through image comparison |
| US9922238B2 (en) * | 2015-06-25 | 2018-03-20 | West Virginia University | Apparatuses, systems, and methods for confirming identity |
| CN108062544A (en) * | 2018-01-19 | 2018-05-22 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of face In vivo detection |
| CN108510583A (en) * | 2018-04-03 | 2018-09-07 | 北京华捷艾米科技有限公司 | The generation method of facial image and the generating means of facial image |
| CN108615256A (en) * | 2018-03-29 | 2018-10-02 | 西南民族大学 | A kind of face three-dimensional rebuilding method and device |
-
2018
- 2018-12-29 CN CN201811638462.0A patent/CN109508702A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101839692A (en) * | 2010-05-27 | 2010-09-22 | 西安交通大学 | Method for measuring three-dimensional position and stance of object with single camera |
| CN101908140A (en) * | 2010-07-29 | 2010-12-08 | 中山大学 | Biopsy method for use in human face identification |
| CN102594964A (en) * | 2012-02-13 | 2012-07-18 | 广东步步高电子工业有限公司 | Intelligent contact adding method using mobile terminal |
| CN103824278A (en) * | 2013-12-10 | 2014-05-28 | 清华大学 | Monitoring camera calibration method and system |
| CN105205438A (en) * | 2014-09-05 | 2015-12-30 | 北京七鑫易维信息技术有限公司 | Method of using infrared eyeball to track and control distance of eyes and screen and system thereof |
| CN105868677A (en) * | 2015-01-19 | 2016-08-17 | 阿里巴巴集团控股有限公司 | Live human face detection method and device |
| US9922238B2 (en) * | 2015-06-25 | 2018-03-20 | West Virginia University | Apparatuses, systems, and methods for confirming identity |
| CN105260726A (en) * | 2015-11-11 | 2016-01-20 | 杭州海量信息技术有限公司 | Interactive video in vivo detection method based on face attitude control and system thereof |
| CN105740781A (en) * | 2016-01-25 | 2016-07-06 | 北京天诚盛业科技有限公司 | Three-dimensional human face in-vivo detection method and device |
| CN106529414A (en) * | 2016-10-14 | 2017-03-22 | 国政通科技股份有限公司 | Method for realizing result authentication through image comparison |
| CN108062544A (en) * | 2018-01-19 | 2018-05-22 | 百度在线网络技术（北京）有限公司 | For the method and apparatus of face In vivo detection |
| CN108615256A (en) * | 2018-03-29 | 2018-10-02 | 西南民族大学 | A kind of face three-dimensional rebuilding method and device |
| CN108510583A (en) * | 2018-04-03 | 2018-09-07 | 北京华捷艾米科技有限公司 | The generation method of facial image and the generating means of facial image |
Non-Patent Citations (4)
| Title |
|---|
| WANG LITING ET AL: "Face Live Detection Method Based on Physiological Motion Analysis", 《TSINGHUA SCIENCE & TECHNOLOGY》 * |
| 张三喜: "《高速摄像及其应用技术》", 31 January 2006 * |
| 张高铭 等: "基于人脸的活体检测系统", 《计算机系统应用》 * |
| 王冬: "基于单目视觉的三维重建技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111860057A (en) * | 2019-04-29 | 2020-10-30 | 北京眼神智能科技有限公司 | Face image blurring and living body detection method and device, storage medium and equipment |
| CN110222573A (en) * | 2019-05-07 | 2019-09-10 | 平安科技（深圳）有限公司 | Face identification method, device, computer equipment and storage medium |
| CN110222573B (en) * | 2019-05-07 | 2024-05-28 | 平安科技（深圳）有限公司 | Face recognition method, device, computer equipment and storage medium |
| CN110164013A (en) * | 2019-05-31 | 2019-08-23 | 广州软盈科技有限公司 | A kind of retrospective caller management method |
| CN110544335A (en) * | 2019-08-30 | 2019-12-06 | 北京市商汤科技开发有限公司 | Object recognition system and method, electronic device, and storage medium |
| CN112711968A (en) * | 2019-10-24 | 2021-04-27 | 浙江舜宇智能光学技术有限公司 | Face living body detection method and system |
| CN111339958A (en) * | 2020-02-28 | 2020-06-26 | 山东笛卡尔智能科技有限公司 | Monocular vision-based face in-vivo detection method and system |
| CN111339958B (en) * | 2020-02-28 | 2023-08-29 | 南京鑫之派智能科技有限公司 | Face living body detection method and system based on monocular vision |
| CN111783599B (en) * | 2020-06-24 | 2023-08-18 | 东风汽车有限公司 | Face living body identification method and electronic equipment |
| CN111783599A (en) * | 2020-06-24 | 2020-10-16 | 东风汽车有限公司 | Face living body identification method and electronic equipment |
| CN112016437A (en) * | 2020-08-26 | 2020-12-01 | 中国科学院重庆绿色智能技术研究院 | Living body detection method based on face video key frame |
| CN112016437B (en) * | 2020-08-26 | 2023-02-10 | 中国科学院重庆绿色智能技术研究院 | Living body detection method based on face video key frame |
| CN112418036A (en) * | 2020-11-12 | 2021-02-26 | 广州市保伦电子有限公司 | Video conference order analysis method and processing terminal |
| CN112487921A (en) * | 2020-11-25 | 2021-03-12 | 奥比中光科技集团股份有限公司 | Face image preprocessing method and system for living body detection |
| CN112487921B (en) * | 2020-11-25 | 2023-09-08 | 奥比中光科技集团股份有限公司 | Face image preprocessing method and system for living body detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109508702A (en) | A kind of three-dimensional face biopsy method based on single image acquisition equipment | |
| US11915365B2 (en) | 3D body model generation | |
| Varol et al. | Learning from synthetic humans | |
| CN111602140B (en) | Method of analyzing objects in images recorded by a camera of a head-mounted device | |
| US8620038B2 (en) | Method, system and computer program product for automatic and semi-automatic modification of digital images of faces | |
| CN109508681A (en) | The method and apparatus for generating human body critical point detection model | |
| US8265351B2 (en) | Method, system and computer program product for automatic and semi-automatic modification of digital images of faces | |
| CN109583285A (en) | Object identifying method | |
| Choppin et al. | The potential of the Microsoft Kinect in sports analysis and biomechanics | |
| US20180184062A1 (en) | Image Processing to Determine Center of Balance in a Digital Image | |
| JP7339386B2 (en) | Eye-tracking method, eye-tracking device, terminal device, computer-readable storage medium and computer program | |
| CN108230293A (en) | Determine method and apparatus, electronic equipment and the computer storage media of quality of human face image | |
| CN108197618A (en) | For generating the method and apparatus of Face datection model | |
| CN109615664B (en) | Calibration method and device for optical perspective augmented reality display | |
| US20220026276A1 (en) | System and Method for Creation of Topical Agents with Improved Image Capture | |
| KR101181967B1 (en) | 3D street view system using identification information. | |
| CN109299658A (en) | Face area detecting method, face image rendering method, device and storage medium | |
| US11798299B2 (en) | Methods and systems for generating 3D datasets to train deep learning networks for measurements estimation | |
| CN109961472B (en) | Method, system, storage medium and electronic device for generating 3D thermodynamic diagram | |
| CN109859857A (en) | Mask method, device and the computer readable storage medium of identity information | |
| CN109859100A (en) | Display methods, electronic equipment and the computer readable storage medium of virtual background | |
| US20220198696A1 (en) | System for determining body measurement from images | |
| JP2019175321A (en) | Image evaluation device, image evaluation method, and computer program | |
| CN105814604B (en) | For providing location information or mobile message with the method and system of at least one function for controlling vehicle | |
| Yin et al. | Depth maps restoration for human using RealSense |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190322 |
|
| RJ01 | Rejection of invention patent application after publication |