CN108876891A - Face image data acquisition method and face image data acquisition device - Google Patents
Face image data acquisition method and face image data acquisition device Download PDFInfo
- Publication number
- CN108876891A CN108876891A CN201711204796.2A CN201711204796A CN108876891A CN 108876891 A CN108876891 A CN 108876891A CN 201711204796 A CN201711204796 A CN 201711204796A CN 108876891 A CN108876891 A CN 108876891A
- Authority
- CN
- China
- Prior art keywords
- acquisition
- scene
- face image
- image data
- dummy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/005—General purpose rendering architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2215/00—Indexing scheme for image rendering
- G06T2215/16—Using real world measurements to influence rendering
Landscapes
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
- Image Processing (AREA)
Abstract
A kind of face image data acquisition method and face image data acquisition device.The facial image collecting method includes：Obtain the dummy acquisition scene for indicating really to acquire scene；Obtain at least one dummy head model data；At least one dummy head model data is respectively put into dummy acquisition scene and is rendered, to obtain at least one face image data.The facial image collecting method makes collecting work be no longer limited by the physical location of gathered person, so as to reduce resource and time consumed by collecting work, and then can be with save the cost.
Description
Technical field
The embodiment of the present disclosure is related to a kind of face image data acquisition method and face image data acquisition device.
Background technique
The one kind of face image data as image data, is widely used in computer vision, artificial intelligence etc. at present
Field, for promoting the performance of such as detection of human face detection and tracing, face character, face critical point detection.Facial image is adopted
Collecting system build and data acquisition needs to expend a large amount of resource and time in itself, and be limited to the physical bit of gathered person
It sets.So how to be virtualized to true man face image acquiring system, becomes one and critically important and urgently to be resolved ask
Topic.
Summary of the invention
A disclosure at least embodiment provides a kind of face image data acquisition method, including：It obtains for indicating true
Acquire the dummy acquisition scene of scene；Obtain at least one dummy head model data；It will at least one described dummy head mould
Type data are respectively put into the dummy acquisition scene and are rendered, to obtain at least one face image data.
For example, the acquisition is for indicating true in the face image data acquisition method that one embodiment of the disclosure provides
The dummy acquisition scene of acquisition scene includes in fact：The inner parameter of the image collecting device in the true acquisition scene is obtained,
And location parameter of the described image acquisition device in the true acquisition scene；Obtain the environment of the true acquisition scene
Information；According to it is described it is true acquisition scene environmental information and described image acquisition device inner parameter and location parameter,
The initialization of virtual for establishing the corresponding true acquisition scene acquires scene；School is carried out to initialization of virtual acquisition scene
Standard obtains the dummy acquisition scene.
For example, in the face image data acquisition method that one embodiment of the disclosure provides, it is described really to be adopted according to described
Collect the environmental information of scene and the inner parameter and location parameter of described image acquisition device, establishes the corresponding true acquisition
The initialization of virtual of scene acquires scene：It is virtual according to the inner parameter of described image acquisition device and location parameter setting
Image collecting device；Virtual environment is set according to the environmental information of the true acquisition scene, obtains the initialization of virtual and adopts
Collect scene.
For example, in the face image data acquisition method that one embodiment of the disclosure provides, it is described empty to the initialization
Quasi- acquisition scene is calibrated, and obtaining the dummy acquisition scene includes：Obtain the acquisition of described image acquisition device is located at institute
State the respective real human face image data of at least one gathered person in true acquisition scene；Obtain at least one described quilt
The respective head model data of picker；The respective head model data of at least one described gathered person are put into respectively described
It is rendered in initialization of virtual acquisition scene, and adjusts the luminance parameter of the initialization of virtual acquisition scene, until rendering
The face distribution of color error of each face image data obtained and corresponding real human face image data is default less than one
Value, obtains the dummy acquisition scene.
For example, the acquisition is described really to adopt in the face image data acquisition method that one embodiment of the disclosure provides
Collection scene environmental information include：Obtain light source information, ambient lighting information and the background information in the true acquisition scene
At least one of.
For example, the light source information includes light in the face image data acquisition method that one embodiment of the disclosure provides
At least one of source location information and light-source angle information.
For example, obtaining the light source position letter in the face image data acquisition method that one embodiment of the disclosure provides
Breath or light-source angle information include：It is described using the head center for the gathered person being located in the true acquisition scene as origin
The head of gathered person and the line direction of described image acquisition device are horizontal axis, and vertical direction is the longitudinal axis, are established three-dimensional
Coordinate system；Position or angle of the light source in the three-dimensional system of coordinate are measured, to obtain the light source position information or the light
Source angle information.
For example, in the face image data acquisition method that one embodiment of the disclosure provides, it is described to obtain the environment light
Include according to information：The panorama diagram data of the true acquisition scene is obtained, and according to the brightness data in the panorama diagram data
Obtain the ambient lighting information.
For example, in the face image data acquisition method that one embodiment of the disclosure provides, it is described to obtain the background letter
Breath includes：The background image of the true acquisition scene of described image acquisition device acquisition is obtained as background information；Alternatively,
The panorama diagram data of the true acquisition scene is obtained, and the background information is obtained according to the panorama diagram data.
A disclosure at least embodiment also provides a kind of face image data acquisition device, including acquiring unit and processing list
Member.The acquiring unit is configured to obtain the dummy acquisition scene for indicating really to acquire scene, and obtains at least one
Dummy head model data.The processing unit is configured at least one described dummy head model data being respectively put into described
It is rendered in dummy acquisition scene, to obtain at least one face image data.
For example, the acquiring unit also configures in the face image data acquisition device that one embodiment of the disclosure provides
It is really adopted to obtain inner parameter and the described image acquisition device of the image collecting device in true acquisition scene described
Collect the location parameter in scene, and obtains the environmental information of the true acquisition scene.The processing unit is additionally configured to basis
The environmental information of the true acquisition scene and the inner parameter and location parameter of described image acquisition device, establish corresponding institute
The initialization of virtual acquisition scene of true acquisition scene is stated, and initialization of virtual acquisition scene is calibrated, is obtained
The dummy acquisition scene.
For example, the processing unit is being established in the face image data acquisition device that one embodiment of the disclosure provides
It is additionally configured to when the initialization of virtual acquisition scene：It is arranged according to the inner parameter of described image acquisition device and location parameter
Virtual image acquisition device；And virtual environment is arranged according to the environmental information of the true acquisition scene, it obtains described initial
Change dummy acquisition scene.
For example, the acquiring unit also configures in the face image data acquisition device that one embodiment of the disclosure provides
For obtain described image acquisition device acquisition be located at it is described it is true acquisition scene at least one gathered person it is respective
Real human face image data, and obtain the respective head model data of at least one described gathered person.The processing unit
It is additionally configured to that the respective head model data of at least one described gathered person are put into initialization of virtual acquisition field respectively
It is rendered in scape, and adjusts the luminance parameter of the initialization of virtual acquisition scene, until rendering each face obtained
The face distribution of color error of image data and corresponding real human face image data obtains described virtually adopting less than a preset value
Collect scene.
A disclosure at least embodiment also provides a kind of storage medium, and being stored on the storage medium may be adapted to processor
The computer instruction of execution, and when computer instruction is executed by the processor, implements the following：It obtains for indicating
The dummy acquisition scene of true acquisition scene；Obtain at least one dummy head model data；It will at least one described virtual head
Portion's model data is respectively put into the dummy acquisition scene and is rendered, to obtain at least one face image data.
A disclosure at least embodiment also provides a kind of face image data acquisition device, including processor and storage are situated between
Matter.The storage medium, which is configured to be stored with, may be adapted to the computer instruction that the processor executes, and the computer instruction
It is implemented the following when being executed by the processor：Obtain the dummy acquisition scene for indicating really to acquire scene；It obtains extremely
A few dummy head model data；At least one described dummy head model data is respectively put into the dummy acquisition scene
In rendered, to obtain at least one face image data.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the embodiment of the present disclosure, the attached drawing to embodiment is simply situated between below
It continues, it should be apparent that, the accompanying drawings in the following description merely relates to some embodiments of the present disclosure, rather than the limitation to the disclosure.
Fig. 1 is a kind of schematic diagram for face image data acquisition method that one embodiment of the disclosure provides；
Fig. 2 is a kind of exemplary schematic diagram of method and step S100 shown in Fig. 1；
Fig. 3 is a kind of exemplary schematic diagram of method and step S130 shown in Fig. 2；
Fig. 4 is a kind of exemplary schematic diagram of method and step S140 shown in Fig. 2；
Fig. 5 is the schematic diagram for the three-dimensional system of coordinate established in embodiment of the disclosure；
Fig. 6 is a kind of schematic diagram of the specific example for face image data acquisition method that one embodiment of the disclosure provides；
Fig. 7 is a kind of schematic diagram for face image data acquisition device that one embodiment of the disclosure provides；
Fig. 8 is a kind of schematic diagram for storage medium that one embodiment of the disclosure provides；And
Fig. 9 is the schematic diagram for another face image data acquisition device that one embodiment of the disclosure provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present disclosure clearer, below in conjunction with the embodiment of the present disclosure
Attached drawing, the technical solution of the embodiment of the present disclosure is described.Obviously, described embodiment is a part of this disclosure reality
Example is applied, instead of all the embodiments.Based on described embodiment of the disclosure, those of ordinary skill in the art are without wound
Every other embodiment obtained under the premise of the property made is worked belongs to the range of disclosure protection.
Unless otherwise defined, the technical term or scientific term that the disclosure uses should be tool in disclosure fields
The ordinary meaning for thering is the personage of general technical ability to be understood." first ", " second " used in the disclosure and similar word are simultaneously
Any sequence, quantity or importance are not indicated, and are used only to distinguish different component parts.Equally, "one", " one " or
The similar word such as person's "the" does not indicate that quantity limits yet, but indicates that there are at least one." comprising " or "comprising" etc. are similar
Word mean to occur element or object before the word cover the element for appearing in the word presented hereinafter or object and its
It is equivalent, and it is not excluded for other elements or object.The similar word such as " connection " or " connected " be not limited to physics or
The connection of person's machinery, but may include electrical connection, it is either direct or indirect."upper", "lower", " left side ",
" right side " etc. is only used for indicating relative positional relationship, after the absolute position for being described object changes, then the relative positional relationship
May correspondingly it change.
The one kind of face image data as image data is widely used in such as computer vision, artificial intelligence at present
The fields such as energy, for promoting the performance of such as detection of human face detection and tracing, face character, face critical point detection.It obtains
A large amount of face image data needs to build man face image acquiring system and carries out data collection task, these can all be expended
A large amount of resource and time.For example, in order to enrich the diversity of face image data, it usually needs really adopted in many different
At this moment the facial image for collecting a large amount of different gathered persons of acquisition in scene just needs each gathered person to go to respectively each true
It acquires in scene, then acquires corresponding face image data in the true acquisition scene.For example, if necessary to obtain 100
The face image data of 100 gathered persons, then need to carry out 100*100 collecting work in a true acquisition scene.Institute
How to be virtualized to true man face image acquiring system, so that collecting work is no longer limited by the object of gathered person
Position is managed, an important and urgent problem to be solved is become.
A disclosure at least embodiment provides a kind of face image data acquisition method, including：It obtains for indicating true
Acquire the dummy acquisition scene of scene；Obtain at least one dummy head model data；By at least one dummy head pattern number
It is rendered according to being respectively put into dummy acquisition scene, to obtain at least one face image data.
A disclosure at least embodiment also provides a kind of face image data acquisition device.
The face image data acquisition method and face image data acquisition device that embodiment of the disclosure provides, can be with
So that man face image acquiring work is no longer limited by the physical location of gathered person, so as to reduce consumed by collecting work
Resource and time, and then can be possibly realized with save the cost but also providing a large amount of face image datas.
Embodiment of the disclosure is described in detail with reference to the accompanying drawing.
Embodiment of the disclosure provides a kind of face image data acquisition method, as shown in Figure 1, the face image data is adopted
Set method includes following operation.
Step S100：Obtain the dummy acquisition scene for indicating really to acquire scene；
Step S200：Obtain at least one dummy head model data；And
Step S300：At least one dummy head model data is respectively put into dummy acquisition scene and is rendered, with
Obtain at least one face image data.
For example, for step S100 in some embodiments, digital virtual can be carried out to some true acquisition scene
Change to obtain corresponding digital virtual acquisition scene, such as panorama camera device acquisition can be used when carrying out digital virtual
The image of true acquisition scene.In another example in further embodiments, can obtain directly from such as storage medium indicates certain
The dummy acquisition scene of one true acquisition scene, such as the dummy acquisition scene stored in the storage medium is to be obtained ahead of time
's.That is, when executing step S100, if the dummy acquisition scene for indicating true acquisition scene has been obtained ahead of time
And be stored in storage medium, then it only needs to call directly in this step, not need again to the true acquisition scene
Carry out virtualization operations.
It should be noted that the true acquisition scene expression one is specific true in embodiment of the disclosure
Scene is acquired, specific true acquisition scene was indicated in a specific time, a specific place and specific here
Light environment under, the scene that is shot using image collecting device towards certain angle.For example, a true acquisition scene
It can be：In the scene that certain office doorway 10 AM is shot towards Due South.In another example a true acquisition scene can
To be：Scene of the 5 PM towards positive north shooting before the Temple of Heaven of Beijing.In embodiment of the disclosure, scene packet is really acquired
Indoor scene is included, also includes outdoor scene, embodiment of the disclosure is not construed as limiting this.
For example, step S200 in some embodiments, can be carried out head model at least one gathered person and be adopted
Collection, to obtain at least one dummy head model data.It, can be directly from for example storing in another example in further embodiments
At least one the dummy head model stored at least one dummy head model data, such as the storage medium is obtained in medium
Data are obtained ahead of time.That is, when executing step S200, if at least one dummy head has been obtained ahead of time
Model data has simultaneously been stored in storage medium, then only needs to call directly in this step, is not needed again at least one
A gathered person carries out head model acquisition.
For example, when carrying out head model acquisition to gathered person, it in some embodiments, can be to the head of gathered person
Laser scanning is carried out to obtain corresponding dummy head model data.In another example can using infrared lens, dot matrix projector come
Dummy head model data is obtained, it is (non-visible in human face's formation multiple infrared light spots for example up to ten thousand by dot matrix projector
Light), infrared camera shooting is recycled, the dot array data obtained using shooting carries out three-dimensional modeling to the face mask scanned.
In another example in further embodiments, can first obtain the head image of the multi-angle of gathered person, three-dimensional reconstruction is then used
Algorithm obtains corresponding dummy head model data.The three-dimensional reconstruction algorithm used in embodiment of the disclosure can use existing
Method, which is not described herein again.
In another example in step s 200, can also directly simulate void using analog computing device and in conjunction with simulation algorithm
Quasi- head model data, as long as the algorithm can approach true people.Such as the analog computing device can be implemented as
The universal computing devices such as computer, also can be implemented as dedicated computing equipment, and embodiment of the disclosure is not construed as limiting this.
For example, in step S300, it can be by least one dummy head model data obtained in step s 200 point
It is not put into the dummy acquisition scene obtained in the step s 100 and is rendered, to obtain at least one face image data.Example
Such as, dummy acquisition scene can establish in rendering engine.Such as rendering engine can use the illusory engine of Unreal3D, at this
The rendering engine of use is not construed as limiting in disclosed embodiment, establishes 3D rendering contexts as long as can be achieved on.
As described above, for example, if needing to obtain the facial image number of 100 gathered persons in 100 true acquisition scenes
According to can obtain respectively for indicating this 100 using the face image data acquisition method provided in embodiment of the disclosure
The dummy acquisition scene of true acquisition scene, and the head model data of 100 gathered persons are obtained respectively, finally should
100 head model datas are respectively put into 100 dummy acquisition scenes and are rendered respectively, to obtain 100*100 face
Image data.The face image data acquisition method provided using the embodiment of the present disclosure can make collecting work no longer limited
In the physical location of gathered person, so as to reduce resource and time consumed by collecting work, and then can save into
This.
In embodiment of the disclosure, head model data include the 3D shape on head, and can also further be wrapped
Include the texture on head.The color of each Minimum Area unit of texture representation face area, plane can be launched by being one
It include the picture of colouring information.Texture can show skin detail such as pore, wrinkle, lip etc..
It should be noted that in embodiment of the disclosure, dummy head model data is indicated when carrying out dummy acquisition
The head model data being put into dummy acquisition scene.
For example, as shown in Fig. 2, in one example, above-mentioned steps S100 may include operating as follows.
Step S110：Obtain the inner parameter and image collecting device of the image collecting device in true acquisition scene
Location parameter in really acquisition scene；
Step S120：Obtain the environmental information of true acquisition scene；
Step S130：According to the environmental information of true acquisition scene and the inner parameter and position ginseng of image collecting device
Number, the initialization of virtual for establishing corresponding true acquisition scene acquire scene；And
Step S140：Initialization of virtual acquisition scene is calibrated, dummy acquisition scene is obtained.
For example, in step s 110, in the case of image collecting device uses camera, the inside ginseng of image collecting device
Numerical example such as may include：Aperture, exposure, time for exposure, focal length, the distance of focal plane to camera and internal photosensitive element are big
It is small etc..For example, inner parameter can also include other parameters, this public affairs in the case of image collecting device uses other equipment
The embodiment opened is not construed as limiting this.
For example, the inner parameter of image collecting device can be stored in advance in storage medium, when executing step S110
It only needs directly to call the inner parameter from storage medium.
In embodiment of the disclosure, image collecting device includes but is not limited to the equipment such as camera, video camera, as long as can
To acquire the image in true acquisition scene (for example including the background image etc. in facial image, true acquisition scene).
For some true acquisition scene, a three-dimensional system of coordinate can be artificially established.For example, as shown in figure 5, can
Using using the head center for really acquiring a gathered person 10 in scene as origin O, head and the image of gathered person 10 is adopted
The line direction of acquisition means 20 is X-axis (horizontal axis), and vertical direction is Y-axis (longitudinal axis), in the horizontal plane the side vertical with X-axis
To for Z axis, a three-dimensional system of coordinate is established.For example, being left-handed coordinate system shown in Fig. 5.It should be noted that the reality of the disclosure
It applies example to be not limited to establish the mode of three-dimensional system of coordinate shown in Fig. 5, such as cylindrical-coordinate system, spheric coordinate system can also be used
Deng.
For example, on the basis of above-mentioned three-dimensional system of coordinate, when executing step S110, such as tape measure, tape measure can be used
Equal measuring tools directly measure the location parameter for obtaining image collecting device in really acquisition scene, can also be passed using distance
The modes measurement such as sensor obtains, and then can also pass through image of shooting itself using the reference scale in image and be calculated.
For example, in one example, step S120 includes obtaining the true light source information acquired in scene, ambient lighting letter
At least one of breath and background information.For example, can only acquire the part in true acquisition scene when executing step S120
Environmental information, such as light source information, and other environmental information such as ambient lighting information can use default value or preset value.
It should be noted that described light source includes lamp and artificial light source in embodiment of the disclosure.Example
Such as lamp includes sunlight.Such as artificial light source includes incandescent lamp, LED light etc..For example, in an indoor scene
In, when opening incandescent lamp and having sunlight to come in from window transmission, light source includes sunlight and incandescent lamp at this time.In another example
In an outdoor scene, when not having using artificial lighting, light source only includes sunlight at this time.
For example, light source information includes at least one of light source position information and light-source angle information.Such as it is built above-mentioned
It, can be by position and angle of the measurement light source in three-dimensional system of coordinate, to obtain light on the basis of vertical three-dimensional system of coordinate
Source location information or light-source angle information.
For example, at this moment only needing to obtain light source position information, such as at one when light source only includes artificial light source
In indoor scene, in the work of only one incandescent lamp, it is only necessary on the basis of the three-dimensional system of coordinate of above-mentioned foundation and use
Measuring tool measures the location information that incandescent lamp can be obtained.For example, at this moment only being needed when light source only includes lamp
Light-source angle information is obtained, such as in an indoor or outdoors scene, when there is no artificial lighting there was only sunlight,
It only needs to measure on the basis of the three-dimensional system of coordinate of above-mentioned foundation and using measuring tool and light-source angle letter can be obtained
Breath, such as light-source angle information can be measured according to the projection of object.
For example, in one example, above-mentioned acquisition really acquires the ambient lighting information in scene and includes：Acquisition is really adopted
Collect the panorama diagram data of scene, and ambient lighting information is obtained according to the brightness data in panorama diagram data.For example, can use
Wide angle camera is shot towards the positive negative direction of X-axis, Y-axis and Z axis respectively centered on the origin O shown in Fig. 5, such as is shot
Six images, shooting result can recorde as 16 RAW files, and synthesize panorama sketch using Panoramic Image Mosaic Algorithm.By institute
It include brightness data in the panorama diagram data of acquisition, it is possible to which true acquisition is obtained according to the brightness data in panorama diagram data
The ambient lighting information of scene.
It should be noted that in embodiment of the disclosure, quantity and record text to the image of wide angle camera shooting
The format of part is not construed as limiting.In addition, ambient lighting information described in embodiment of the disclosure indicates in true acquisition scene
Environment luminance information, such as illumination provided by incandescent lamp or sunlight really acquisition scene in pass through multiple diffusing reflection
The brightness at the different location in environment can be made different.The ambient lighting information of above-mentioned acquisition is to characterize true acquisition field
The brightness at different location in scape.
In addition, the panorama diagram data of above-mentioned true acquisition scene can also be stored in advance in such as storage medium, needing
It is called directly when wanting.
For example, in one example, the background information that above-mentioned acquisition really acquires in scene includes：Obtain image collector
The background image of true acquisition scene of acquisition is set as background information.For example, needing acquisition to be located at when executing step S140
The real human face image data of gathered person in true acquisition scene, such as shown in figure 5, existed using image collecting device 20
The real human face image data of gathered person is acquired under conditions of certain location parameter and inner parameter.When acquisition is completed
Afterwards, gathered person can be removed from true acquisition scene, in the location parameter that does not change image collecting device 20 and interior
In the case of portion's parameter, the background image of true acquisition scene is acquired as background information.It should be noted that above-mentioned really adopt
The background image of collection scene can also be stored in advance in such as storage medium, be called directly when needed.
For example, in another example, above-mentioned acquisition, which really acquires the background information in scene, to include：It obtains true
The panorama diagram data of real acquisition scene, and background information is obtained according to panorama diagram data.Due to having included in panorama diagram data
The image data of six direction is total to towards the positive negative direction of X-axis, Y-axis and Z axis in true acquisition scene, it is possible to from panorama
Background image of the image collecting device towards gathered person direction is extracted in diagram data.It can on how to obtain panorama diagram data
With with reference to corresponding description among the above, which is not described herein again.
For example, as shown in figure 3, in one example, step S130 may include operating as follows.
Step S131：According to the inner parameter of image collecting device and location parameter, virtual image acquisition device is set；With
And
Step S132：Virtual environment is set according to the environmental information of true acquisition scene, obtains initialization of virtual acquisition field
Scape.
For example, in step s 130, initialization of virtual acquisition scene can be established in a rendering engine.Such as it renders
Engine can use the illusory engine of Unreal3D, be not construed as limiting in embodiment of the disclosure to the rendering engine of use, as long as
It can be achieved on and establish 3D rendering contexts.
For example, in step S131, it can be in rendering engine according to the image collecting device obtained in step S110
Virtual image acquisition device is arranged in inner parameter and location parameter, that is to say, that simulated in rendering engine come one with
The similar virtual image acquisition device of true image collecting device, in subsequent progress dummy acquisition work.
For example, in step S132, it can be in rendering engine according in the true acquisition scene obtained in step S120
Environmental information be arranged virtual environment.For example, can be believed in rendering engine according to the light source in the true acquisition scene of acquisition
Virtual light source in breath setting virtual environment；According to the ambient lighting information of the true acquisition scene of acquisition, virtual environment is set
Ambient lighting；The background of virtual environment is set according to the background information of the true acquisition scene of acquisition.
For example, as shown in figure 4, in one example, step S140 may include operating as follows.
Step S141：Obtain at least one gathered person's of image acquisition device being located in true acquisition scene
Respective real human face image data；
Step S142：Obtain the respective head model data of at least one gathered person；And
Step S143：The respective head model data of at least one gathered person are put into initialization of virtual acquisition field respectively
It is rendered in scape, and adjusts the luminance parameter of initialization of virtual acquisition scene, until rendering each facial image obtained
The face distribution of color error of data and corresponding real human face image data obtains dummy acquisition scene less than a preset value.
For example, in step s 141, such as really adopt as shown in figure 5, can be located at using the acquisition of image collecting device 20
Collect the respective real human face image data of at least one gathered person 10 in scene.It should be noted that in step S141
Real human face image data be also possible to be stored in advance in such as storage medium, call directly when needed.
In step S142, the respective head model data of at least one gathered person in obtaining step 141.Example
Such as, head model acquisition can be carried out by the respective head at least one gathered person to obtain head model data.
In another example the respective head model data of at least one gathered person can be directly obtained from such as storage medium, such as
The respective head model data of at least one gathered person stored in the storage medium are that preparatory acquisition obtains.
It, can be by the respective head model number of at least one gathered person obtained in step S142 in step S143
It is rendered according to being put into the initialization of virtual established in step s 130 acquisition scene, can be obtained and be corresponded to often by rendering
The face image data of one head model data.Then the brightness of initialization of virtual acquisition scene in rendering engine is gradually adjusted
Parameter (obtains in step s 141 until rendering each face image data obtained and corresponding real human face image data
Take) face distribution of color error less than a preset value, to obtain dummy acquisition scene.
For example, being carried out in step S143 to the obtained face image data of rendering and corresponding real human face image data
When comparison, the error sum of squares of individual element can be made less than a preset value, such as relative to the percentage of a certain reference value.Example
Such as, which can be specifically arranged according to the actual situation, and under different true acquisition scenes, which can be different.
It should be noted that one can be used when calibrating in step S140 to initialization of virtual acquisition scene
The real human face image data and head model data of gathered person is calibrated, and two, three or more can also be used
The real human face image of gathered person and respective head model data are calibrated, and embodiment of the disclosure does not limit this
It is fixed.It is calibrated relative to the data only with a gathered person, using the data of multiple gathered persons to calibrate can be with
Improve the accuracy for the dummy acquisition scene established.
In an example of the disclosure, as shown in fig. 6, providing a kind of specifically showing for face image data acquisition method
Example.This method includes operating procedure shown in Fig. 6.
For example, first using true acquisition system (for example including image collecting device, head model acquisition device, measurement
Tool etc.) execute step S11, S12, S13, S14, S15 and S16.
For example, in step s 11, it can be using the true people of gathered person of the camera acquisition in true acquisition scene
Face image data.In step s 12, gathered person is joined after removing in true acquisition scene in the inside for not changing camera
In the case of several and location parameter, the background image of true acquisition scene is acquired.In step s 13, measuring tool can be used
Measure location parameter of the camera in really acquisition scene.It, can be using the position of measuring tool measurement light source in step S14
Or angle information.It, can be using the panorama sketch of the true acquisition scene of wide angle camera acquisition in step S15.In step s 16,
It can be can wrap using the head model data of head model acquisition device acquisition gathered person, such as head model acquisition device
Include 3D Laser Scanning Equipment, matrix camera etc..
For example, in step S17, it can be according to the data obtained in step S12, S13, S14, S15 in rendering engine
Establish initialization of virtual acquisition scene.It, can be according to step S11 in step S18 after establishing initialization of virtual acquisition scene
Initialization of virtual acquisition scene is calibrated with the data obtained in S16, so as to obtain dummy acquisition scene.
For example, the dummy head model data obtained in step S19 can be placed in step S18 in step S20
Obtained in dummy acquisition scene, by rendering to obtain face image data.
The face image data acquisition method provided using the embodiment of the present disclosure, can make collecting work be no longer limited by
The physical location of gathered person, so as to reduce resource and time consumed by collecting work, and then can be with save the cost.
One embodiment of the disclosure also provides a kind of face image data acquisition device 10, as shown in fig. 7, the face figure
As data acquisition device 10 includes acquiring unit 100 and processing unit 200.
For example, the acquiring unit 100 is configured to obtain the dummy acquisition scene for indicating really to acquire scene, and obtain
Take at least one dummy head model data.That is, the acquiring unit 100 is configured to execute the step in above-described embodiment
S100 and step S200.For example, dummy acquisition scene and dummy head model data can be being obtained ahead of time and be stored in
In storage medium, which can directly be called from storage medium.Such as the storage medium can be and be arranged in people
In face image data acquisition device 10 itself, it is also possible to be arranged elsewhere, embodiment of the disclosure does not make this
It limits.
For example, the processing unit 200 is configured at least one dummy head model data being respectively put into dummy acquisition field
It is rendered in scape, to obtain at least one face image data.That is, the processing unit 200 be configured to execute it is above-mentioned
Step S300 in embodiment.For example, the processing unit 200 may include rendering engine, dummy acquisition scene be can establish
In the rendering engine.
Embodiment of the disclosure provide face image data acquisition device can make collecting work be no longer limited by by
The physical location of picker, so as to reduce resource and time consumed by collecting work, and then can be with save the cost.
For example, acquiring unit 100 is additionally configured to obtain in true acquisition scene in an example of the embodiment of the present disclosure
Image collecting device location parameter in really acquisition scene of inner parameter and image collecting device, and obtain true
The environmental information of real acquisition scene.That is, the acquiring unit 100 is also configured as executing the step in above-described embodiment
S110 and step S120.Environmental information about the inner parameter of image collecting device, location parameter and true acquisition scene
It can refer to and accordingly be described in above-described embodiment, which is not described herein again.
Processing unit 200 is additionally configured to the inside ginseng of environmental information and image collecting device according to true acquisition scene
Several and location parameter, the initialization of virtual for establishing corresponding true acquisition scene acquires scene, and acquires field to initialization of virtual
Scape is calibrated, and dummy acquisition scene is obtained.That is, the processing unit 200 is also configured as executing above-described embodiment
In step S130 and step S140.
For example, processing unit 200 is when establishing initialization of virtual acquisition scene in an example of the embodiment of the present disclosure
It is additionally configured to：Virtual image acquisition device is set according to the inner parameter of image collecting device and location parameter, and according to true
Virtual environment is arranged in the environmental information of real acquisition scene, obtains initialization of virtual acquisition scene.That is, the processing unit
200 are also configured as executing the step S131 and step S132 in above-described embodiment.
For example, acquiring unit 100 is additionally configured to acquisition image collecting device and adopts in an example of the embodiment of the present disclosure
The respective real human face image data at least one gathered person of collection being located in true acquisition scene, and obtain at least
The respective head model data of one gathered person.That is, the acquiring unit 100 is also configured as executing above-mentioned implementation
Step S141 and step S142 in example.
Processing unit 200 is additionally configured to that the respective head model data of at least one gathered person are put into initialization respectively
It is rendered in dummy acquisition scene, and adjusts the luminance parameter of initialization of virtual acquisition scene, it is obtained every until rendering
The face distribution of color error of a face image data and corresponding real human face image data obtains virtual less than a preset value
Acquire scene.That is, the processing unit 200 is also configured as executing the step S143 in above-described embodiment.
It should be noted that the acquiring unit in the face image data acquisition device 10 provided in the embodiment of the present disclosure
100 and processing unit 200 can be implemented as including specific integrated circuit, hardware (circuit), firmware or any other combination side
Formula, to realize the function of each unit wishes, such as digital signal processor can be implemented as etc..
Alternatively, the acquiring unit 100 and processing unit 200 in face image data acquisition device 10 also can be implemented as
Including pocessor and storage media, which is configured to be stored with the computer instruction that may be adapted to processor execution, and counts
The function of each unit wishes may be implemented in the instruction of calculation machine when being executed by processor.Embodiment of the disclosure is not construed as limiting this.
One embodiment of the disclosure also provides a kind of storage medium 500, as shown in figure 8, storing on the storage medium 500
There is the computer instruction 501 that may be adapted to processor execution, and following behaviour can be implemented in when computer instruction 501 is executed by processor
Make：
Obtain the dummy acquisition scene for indicating really to acquire scene；
Obtain at least one dummy head model data；And
At least one dummy head model data is respectively put into dummy acquisition scene and is rendered, to obtain at least one
A face image data.
That is, processor can execute step S100, step S200 and step S300 in above-described embodiment.
For example, in one example, storage medium 500 can be set in computing device, and computing device is configurable to
It is connect with a display device.Such as computing device can also include processor, processor, which can call, is stored in storage medium 500
In computer instruction 501.
One embodiment of the disclosure also provides a kind of face image data acquisition device 10, as shown in figure 9, at the image
Managing device 10 includes processor 600 and storage medium 500.The meter that may be adapted to the execution of processor 600 is stored on storage medium 500
Calculation machine instruction 501, and computer instruction 501 by processor 600 execute when following operation can be implemented：
Obtain the dummy acquisition scene for indicating really to acquire scene；
Obtain at least one dummy head model data；And
At least one dummy head model data is respectively put into dummy acquisition scene and is rendered, to obtain at least one
A face image data.
That is, processor 600 can execute step S100, step S200 and step in above-described embodiment
S300。
In the presently disclosed embodiments, processor can be real by universal integrated circuit chip or dedicated IC chip
It is existing, such as the IC chip can be set on a mainboard, for example, be also provided on the mainboard memory with
And power circuit etc.；In addition, processor can also be by circuit or using software, hardware (circuit), firmware or any combination thereof
Mode is realized.In embodiment of the disclosure, processor may include various calculating structures, such as Complex Instruction Set Computer
(CISC) structure, Reduced Instruction Set Computer (RISC) structure or a kind of structure for carrying out a variety of instruction set combinations.Some
In embodiment, processor is also possible to microprocessor, such as X 86 processor or arm processor, or can be digital processing unit
(DSP) etc..
In embodiment of the disclosure, storage medium for example can be set on above-mentioned mainboard, and storage medium can save
The instruction and/or data that processor executes.For example, storage medium may include one or more computer program products, it is described
Computer program product may include various forms of computer-readable memories, such as volatile memory and/or non-volatile
Property memory.The volatile memory for example may include random access memory (RAM) and/or cache memory
(cache) etc..The nonvolatile memory for example may include read-only memory (ROM), disk, CD, semiconductor storage
Device (such as flash memory etc.) etc..It can store one or more computer program instructions on the computer-readable memory, locate
Reason device can run described program instruction, to realize in the embodiment of the present disclosure and (be realized by processor) desired function.
The above, the only specific embodiment of the disclosure, but the protection scope of the disclosure is not limited thereto, this public affairs
The protection scope opened should be based on the protection scope of the described claims.
Claims (15)
1. a kind of face image data acquisition method, including：
Obtain the dummy acquisition scene for indicating really to acquire scene；
Obtain at least one dummy head model data；
At least one described dummy head model data is respectively put into the dummy acquisition scene and is rendered, to obtain extremely
A few face image data.
2. face image data acquisition method according to claim 1, wherein the acquisition is for indicating really to acquire field
The dummy acquisition scene of scape includes：
The inner parameter and described image acquisition device for obtaining the image collecting device in the true acquisition scene are described
Location parameter in true acquisition scene；
Obtain the environmental information of the true acquisition scene；
According to the environmental information of the true acquisition scene and the inner parameter and location parameter of described image acquisition device, build
The initialization of virtual of the vertical corresponding true acquisition scene acquires scene；
Initialization of virtual acquisition scene is calibrated, the dummy acquisition scene is obtained.
3. face image data acquisition method according to claim 2, wherein described according to the true acquisition scene
The inner parameter and location parameter of environmental information and described image acquisition device establish the first of the corresponding true acquisition scene
Beginningization dummy acquisition scene includes：
According to the inner parameter of described image acquisition device and location parameter, virtual image acquisition device is set；
Virtual environment is set according to the environmental information of the true acquisition scene, obtains the initialization of virtual acquisition scene.
4. face image data acquisition method according to claim 2, wherein described to acquire field to the initialization of virtual
Scape is calibrated, and is obtained the dummy acquisition scene and is included：
Obtain the respective of at least one gathered person of described image acquisition device acquisition being located in the true acquisition scene
Real human face image data；
Obtain the respective head model data of at least one described gathered person；
The respective head model data of at least one described gathered person are put into the initialization of virtual acquisition scene respectively
It is rendered, and adjusts the luminance parameter of the initialization of virtual acquisition scene, until rendering each facial image obtained
The face distribution of color error of data and corresponding real human face image data obtains the dummy acquisition field less than a preset value
Scape.
5. face image data acquisition method according to claim 2, wherein the acquisition true acquisition scene
Environmental information includes：
Obtain at least one of light source information, ambient lighting information and the background information in the true acquisition scene.
6. face image data acquisition method according to claim 5, wherein the light source information includes light source position letter
At least one of breath and light-source angle information.
7. face image data acquisition method according to claim 6, wherein obtain the light source position information or light source
Angle information includes：
Using the head center for the gathered person being located in the true acquisition scene as origin, the head of the gathered person and institute
The line direction for stating image collecting device is horizontal axis, and vertical direction is the longitudinal axis, establishes three-dimensional system of coordinate；
Position or angle of the light source in the three-dimensional system of coordinate are measured, to obtain the light source position information or the light source angle
Spend information.
8. face image data acquisition method according to claim 5, wherein described to obtain the ambient lighting packet
It includes：
The panorama diagram data of the true acquisition scene is obtained, and according to the brightness data acquisition in the panorama diagram data
Ambient lighting information.
9. face image data acquisition method according to claim 5, wherein described to obtain the background information and include：
The background image of the true acquisition scene of described image acquisition device acquisition is obtained as background information；Or
The panorama diagram data of the true acquisition scene is obtained, and the background information is obtained according to the panorama diagram data.
10. a kind of face image data acquisition device, including acquiring unit and processing unit, wherein
The acquiring unit is configured to obtain the dummy acquisition scene for indicating really to acquire scene, and obtains at least one
Dummy head model data；
The processing unit is configured at least one described dummy head model data being respectively put into the dummy acquisition scene
In rendered, to obtain at least one face image data.
11. face image data acquisition device according to claim 10, wherein
The acquiring unit is additionally configured to obtain the inner parameter and the figure of the image collecting device in true acquisition scene
As location parameter of the acquisition device in the true acquisition scene, and obtain the environmental information of the true acquisition scene；
The processing unit is additionally configured to environmental information and described image acquisition device according to the true acquisition scene
Inner parameter and location parameter, the initialization of virtual for establishing the corresponding true acquisition scene acquire scene, and to described first
Beginningization dummy acquisition scene is calibrated, and the dummy acquisition scene is obtained.
12. face image data acquisition device according to claim 11, wherein the processing unit establish it is described just
It is additionally configured to when beginningization dummy acquisition scene：
According to the inner parameter of described image acquisition device and location parameter, virtual image acquisition device is set；And
Virtual environment is set according to the environmental information of the true acquisition scene, obtains the initialization of virtual acquisition scene.
13. face image data acquisition device according to claim 11, wherein
The acquiring unit is additionally configured to obtain being located in the true acquisition scene extremely for described image acquisition device acquisition
The respective real human face image data of a few gathered person, and obtain the respective head of at least one gathered person
Model data；
The processing unit is additionally configured to respectively be put into the respective head model data of at least one described gathered person described
It is rendered in initialization of virtual acquisition scene, and adjusts the luminance parameter of the initialization of virtual acquisition scene, until rendering
The face distribution of color error of each face image data obtained and corresponding real human face image data is default less than one
Value, obtains the dummy acquisition scene.
14. a kind of storage medium, wherein the computer instruction that may be adapted to processor execution is stored on the storage medium, and
The computer instruction implements the following when being executed by the processor：
Obtain the dummy acquisition scene for indicating really to acquire scene；
Obtain at least one dummy head model data；
At least one described dummy head model data is respectively put into the dummy acquisition scene and is rendered, to obtain extremely
A few face image data.
15. a kind of face image data acquisition device, including pocessor and storage media, wherein
The storage medium, which is configured to be stored with, may be adapted to the computer instruction that the processor executes, and the computer instruction
It is implemented the following when being executed by the processor：
Obtain the dummy acquisition scene for indicating really to acquire scene；
Obtain at least one dummy head model data；
At least one described dummy head model data is respectively put into the dummy acquisition scene and is rendered, to obtain extremely
A few face image data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711204796.2A CN108876891B (en) | 2017-11-27 | 2017-11-27 | Face image data acquisition method and face image data acquisition device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711204796.2A CN108876891B (en) | 2017-11-27 | 2017-11-27 | Face image data acquisition method and face image data acquisition device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108876891A true CN108876891A (en) | 2018-11-23 |
| CN108876891B CN108876891B (en) | 2021-12-28 |
Family
ID=64325808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711204796.2A Active CN108876891B (en) | 2017-11-27 | 2017-11-27 | Face image data acquisition method and face image data acquisition device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108876891B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636886A (en) * | 2018-12-19 | 2019-04-16 | 网易（杭州）网络有限公司 | Processing method, device, storage medium and the electronic device of image |
| CN117495664A (en) * | 2023-12-25 | 2024-02-02 | 成都白泽智汇科技有限公司 | Intelligent auxiliary cosmetic system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130120451A1 (en) * | 2011-11-16 | 2013-05-16 | Canon Kabushiki Kaisha | Image processing device, image processing method, and program |
| CN105825544A (en) * | 2015-11-25 | 2016-08-03 | 维沃移动通信有限公司 | Image processing method and mobile terminal |
| CN105850113A (en) * | 2014-01-06 | 2016-08-10 | 欧库勒斯虚拟现实有限责任公司 | Calibration of virtual reality systems |
| CN106127846A (en) * | 2016-06-28 | 2016-11-16 | 乐视控股（北京）有限公司 | Virtual reality terminal and vision virtual method thereof and device |
| CN107193371A (en) * | 2017-04-28 | 2017-09-22 | 上海交通大学 | A kind of real time human-machine interaction system and method based on virtual reality |
| WO2017171005A1 (en) * | 2016-04-01 | 2017-10-05 | 株式会社ｗｉｓｅ | 3-d graphic generation, artificial intelligence verification and learning system, program, and method |
-
2017
- 2017-11-27 CN CN201711204796.2A patent/CN108876891B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130120451A1 (en) * | 2011-11-16 | 2013-05-16 | Canon Kabushiki Kaisha | Image processing device, image processing method, and program |
| CN105850113A (en) * | 2014-01-06 | 2016-08-10 | 欧库勒斯虚拟现实有限责任公司 | Calibration of virtual reality systems |
| CN105825544A (en) * | 2015-11-25 | 2016-08-03 | 维沃移动通信有限公司 | Image processing method and mobile terminal |
| WO2017171005A1 (en) * | 2016-04-01 | 2017-10-05 | 株式会社ｗｉｓｅ | 3-d graphic generation, artificial intelligence verification and learning system, program, and method |
| CN106127846A (en) * | 2016-06-28 | 2016-11-16 | 乐视控股（北京）有限公司 | Virtual reality terminal and vision virtual method thereof and device |
| CN107193371A (en) * | 2017-04-28 | 2017-09-22 | 上海交通大学 | A kind of real time human-machine interaction system and method based on virtual reality |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636886A (en) * | 2018-12-19 | 2019-04-16 | 网易（杭州）网络有限公司 | Processing method, device, storage medium and the electronic device of image |
| US11093733B2 (en) | 2018-12-19 | 2021-08-17 | Netease (Hangzhou) Network Co., Ltd. | Image processing method and apparatus, storage medium and electronic device |
| CN117495664A (en) * | 2023-12-25 | 2024-02-02 | 成都白泽智汇科技有限公司 | Intelligent auxiliary cosmetic system |
| CN117495664B (en) * | 2023-12-25 | 2024-04-09 | 成都白泽智汇科技有限公司 | Intelligent auxiliary cosmetic system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108876891B (en) | 2021-12-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111060023B (en) | High-precision 3D information acquisition equipment and method | |
| US7522163B2 (en) | Method and apparatus for determining offsets of a part from a digital image | |
| CN101292255B (en) | Artifact mitigation in three-dimensional imaging | |
| CN104330074B (en) | Intelligent surveying and mapping platform and realizing method thereof | |
| JP4284664B2 (en) | Three-dimensional shape estimation system and image generation system | |
| JP6344050B2 (en) | Image processing system, image processing apparatus, and program | |
| EP3382645B1 (en) | Method for generation of a 3d model based on structure from motion and photometric stereo of 2d sparse images | |
| CN112016570B (en) | Three-dimensional model generation method for background plate synchronous rotation acquisition | |
| WO2017143745A1 (en) | Method and apparatus for determining movement information of to-be-detected object | |
| JP2003203220A (en) | Three-dimensional image processing method, three- dimensional image processor, there-dimensional image processing system and three-dimensional image processing program | |
| CN112767300A (en) | Method for automatically generating labeling data of hand and method for calculating skeleton length | |
| CN108154551A (en) | The method and system of real-time dynamic reconstruction three-dimensional (3 D) manikin | |
| CN108761994A (en) | The detection method and device of the assembling of camera module and virtual reality device | |
| JP2009081853A (en) | Imaging system and method | |
| CN109559349A (en) | A kind of method and apparatus for calibration | |
| CN110443802B (en) | Image detection method and device, equipment and storage medium | |
| CN113870430B (en) | Workpiece data processing method and device | |
| CN112686084A (en) | Image annotation system | |
| Logothetis et al. | A cnn based approach for the point-light photometric stereo problem | |
| JP2015233266A (en) | Image processing system, information processing device, and program | |
| CN111060008A (en) | 3D intelligent vision equipment | |
| CN108876891A (en) | Face image data acquisition method and face image data acquisition device | |
| CN108648141A (en) | A kind of image split-joint method and device | |
| JPH04256185A (en) | Method for collecting sample picture of picture recognition system | |
| CN101751654A (en) | Application of light information collected in real environment in virtual reality |
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 |