CN106780714A - The generation method and system of face 3D models - Google Patents
The generation method and system of face 3D models Download PDFInfo
- Publication number
- CN106780714A CN106780714A CN201610998968.7A CN201610998968A CN106780714A CN 106780714 A CN106780714 A CN 106780714A CN 201610998968 A CN201610998968 A CN 201610998968A CN 106780714 A CN106780714 A CN 106780714A
- Authority
- CN
- China
- Prior art keywords
- face
- model
- models
- human
- anon
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 84
- 238000012546 transfer Methods 0.000 claims abstract description 34
- 238000000605 extraction Methods 0.000 claims abstract description 32
- 230000011218 segmentation Effects 0.000 claims abstract description 11
- 238000009499 grossing Methods 0.000 claims description 42
- 239000000284 extract Substances 0.000 claims description 10
- 230000001568 sexual effect Effects 0.000 claims description 10
- 230000007547 defect Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000003625 skull Anatomy 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to 3D modelling techniques field, a kind of generation method and system of face 3D models are disclosed.The generation method of the face 3D models, including：Obtain human face scanning model；Split the human face scanning model of acquisition to extract the positive face part of human face scanning model；Human face scanning model according to obtaining judges human body sex；Human body sex according to judging transfers corresponding male/female common model；The common model transferred of segmentation is extracting the anon-normal face part of common model；The anon-normal face part of the positive face part of the human face scanning model of extraction and common model is carried out into stitching processing, to generate face 3D models；And the face 3D models to generating are smoothed, and export the face 3D models.Technical scheme, improves success rate, accuracy and the operating efficiency of face 3D modeling.
Description
Technical field
The present invention relates to 3D modelling techniques field, more particularly to a kind of generation method and system of face 3D models.
Background technology
With computer hardware and software technology and the high speed development of image processing techniques, 3D models start to be applied to all trades and professions
It is central.In human body 3D modeling, human body is generally scanned by 3D body-scanners and body scans model is generated, but this mode
Easily there is mistake or error in anon-normal face part such as hair and clothes in the body scans model of generation, and then causes face 3D
The error rate and error of modeling are higher and success rate is relatively low.It is amendment mistake and error, existing human body 3D modeling generally needs U.S.
Workman's work to generate body scans model carry out landscaping treatment, cause cause face 3D modeling accuracy and operating efficiency compared with
It is low.
The content of the invention
In consideration of it, the present invention provides a kind of generation method and system of face 3D models, when solving existing human body 3D modeling
Cause success rate, accuracy and the relatively low technical problem of operating efficiency because the error rate and error of body scans model are higher.
According to one embodiment of present invention, there is provided a kind of generation method of face 3D models, including：Obtain human face scanning
Model；Split the human face scanning model of acquisition to extract the positive face part of human face scanning model；According to the human face scanning mould for obtaining
Type judges human body sex；Human body sex according to judging transfers corresponding male/female common model；The common model transferred of segmentation with
Extract the anon-normal face part of common model；The positive face part and the anon-normal face part of common model of the human face scanning model that will be extracted
Stitching processing is carried out, to generate face 3D models；And the face 3D models to generating are smoothed, and export the people
Face 3D models.
Preferably, the human face scanning model that the segmentation is obtained extracting the positive face part of human face scanning model, including：Adjust
Take faceform's recognizer；Segmentation portion is carried out to the human face scanning model for obtaining according to the faceform's recognizer transferred
Reason；And recognize and extract the positive face part of human face scanning model.
Preferably, the positive face part of the human face scanning model by extraction and the anon-normal face part of common model are stitched
Conjunction is processed, to generate face 3D models, including：Obtain the positive face part of human face scanning model and the anon-normal face part of common model
Size；The size of the positive face part of the human face scanning model is adjusted, to be adapted to the anon-normal face part of the common model
Size；And the positive face part of the human face scanning model after adjustment is carried out into suture with the anon-normal face part of the common model
Reason, to generate face 3D models.
Preferably, the face 3D models of described pair of generation are smoothed, including：Transfer the smooth calculation of faceform's seam
Method；And according to the faceform's seam smoothing algorithm transferred to generate face 3D models be smoothed with eliminate connect
Seam.
Preferably, the acquisition human face scanning model, including：Face is scanned by 3D body-scanners to be swept to obtain face
Retouch model.
According to another embodiment of the invention, a kind of generation system of face 3D models is also provided, including：Human face scanning
Model acquisition module, for obtaining human face scanning model；Positive face extraction module, mould is obtained for splitting the human face scanning model
The human face scanning model that block is obtained is extracting the positive face part of human face scanning model；Sexual discriminating module, for according to the people
The human face scanning model that face scan model acquisition module is obtained judges human body sex；Common model transfers module, for according to institute
The human body sex for stating the judgement of Sexual discriminating module transfers corresponding male/female common model；Anon-normal face extraction module, for splitting
Common model is stated to transfer the common model of module calls to extract the anon-normal face part of common model；Stitching processing module, is used for
It is public that the positive face part of the human face scanning model that the positive face extraction module is extracted and the anon-normal face extraction module are extracted
The anon-normal face part of model carries out stitching processing, to generate face 3D models；And smoothing processing and output module, for institute
The face 3D models for stating the generation of stitching processing module are smoothed and export the face 3D models.
Preferably, the positive face extraction module includes：Faceform's recognizer transfers unit, for transferring faceform
Recognizer；Dividing processing unit, for transferring faceform's identification that unit is transferred according to faceform's recognizer
Algorithm carries out dividing processing to the human face scanning model that the human face scanning model acquisition module is obtained；And identification and extraction list
Unit, for recognizing and extract the dividing processing unit dividing processing human face scanning model positive face part.
Preferably, the stitching processing module includes：Size acquiring unit, the positive face for obtaining human face scanning model
Divide the size with the anon-normal face part of common model；Size adjusting unit, the positive face for adjusting the human face scanning model
The size divided, to be adapted to the size of the anon-normal face part of the common model；And stitching processing unit, for by the size
The positive face part of human face scanning model after adjustment unit adjustment carries out stitching processing with the anon-normal face part of the common model,
To generate face 3D models.
Preferably, the smoothing processing and output module include：Faceform's seam smoothing algorithm transfers unit, for adjusting
Take faceform's seam smoothing algorithm；Smoothing processing unit, for transferring unit according to faceform's seam smoothing algorithm
The faceform's seam smoothing algorithm transferred is smoothed to eliminate seam to the face 3D models for generating；And output is single
Unit, for exporting the face 3D models after the smoothing processing unit smoothing processing.
Preferably, the human face scanning model acquisition module is 3D body-scanners.
The positive face of human face scanning model is extracted in the generation method and system of the face 3D models that the present invention is provided, respectively segmentation
Part and the anon-normal face part of common model, and carry out stitching processing and smoothing processing to both to generate new face 3D moulds
Type, remains typical positive face characteristic in body scans model, has merged typical anon-normal face features in common model
Point, while influence and the error that sewn seams are caused to face 3D models are also avoid, compared to existing by art designing's artificial treatment
Operating efficiency relatively low during wrong or defect and the relatively low degree of accuracy, effectively prevent existing 3D human bodies and sweep in body scans model
Easily there is mistake and causes the error rate and error of face 3D models in anon-normal face part in the body scans model for retouching instrument generation
Problem higher, while conveniently and efficiently generating accurate face 3D models, improves success rate, the accuracy of face 3D modeling
And operating efficiency.
Brief description of the drawings
Technical scheme in order to illustrate more clearly the embodiments of the present invention, below will be to that will make needed for embodiment description
Accompanying drawing is briefly described.It should be evident that drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is the schematic flow sheet of the generation method of face 3D models in one embodiment of the invention.
Fig. 2 is the schematic flow sheet of the positive face part of segmentation extraction human face scanning model in one embodiment of the invention.
Fig. 3 is the schematic flow sheet of stitching processing generation face 3D models in one embodiment of the invention.
Fig. 4 is the schematic flow sheet of face 3D model smoothings treatment in one embodiment of the invention.
Fig. 5 is the structural representation of the generation system of face 3D models in another embodiment of the present invention.
Fig. 6 is the structural representation of positive face extraction module in another embodiment of the present invention.
Fig. 7 is the structural representation of stitching processing module in another embodiment of the present invention.
Fig. 8 is the structural representation of smoothing processing and output module in another embodiment of the present invention.
Specific embodiment
Make further more detailed description to technical scheme with reference to the accompanying drawings and detailed description.It is aobvious
So, described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based in the present invention
Embodiment, the every other embodiment that those of ordinary skill in the art are obtained on the premise of creative work is not made,
The scope of protection of the invention should all be belonged to.
In the description of the invention, it is to be understood that term " first ", " second " etc. be only used for describe purpose, without
It is understood that to indicate or implying relative importance.In the description of the invention, it is necessary to explanation, specifies unless otherwise clear and definite
And restriction, term " connected ", " connection " should be interpreted broadly, for example, it may be fixedly connected, or be detachably connected,
Or be integrally connected；Can mechanically connect, or electrically connect；Can be joined directly together, it is also possible to by intermediary
It is indirectly connected to.For the ordinary skill in the art, concrete condition can be combined and understands above-mentioned term in the present invention
Concrete meaning.Additionally, in the description of the invention, unless otherwise indicated, " multiple " is meant that two or more.
Any process described otherwise above or method description in flow chart or herein is construed as, and expression includes
It is one or more for realizing specific logical function or process the step of the module of code of executable instruction, fragment or portion
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discussion suitable
Sequence, including function involved by basis by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Fig. 1 is the schematic flow sheet of the generation method of face 3D models in one embodiment of the invention.As illustrated, described
The generation method of face 3D models, including：
Step S101：Obtain human face scanning model.
When needing to set up face 3D models for specific human body, human body can be placed in the electric rotating of 3D body-scanners
Rotated on disk, the face part of the sensor scan human body of the 3D body-scanners, to obtain human face scanning model, the people
Face scan model includes the three-dimensional data of face part.
Step S102：Split the human face scanning model of acquisition to extract the positive face part of human face scanning model.
After human face scanning model is obtained by 3D body-scanners, the positive face part of faceform is extracted.Specifically, ginseng
See Fig. 2, the positive face part of human face scanning model is extracted in the segmentation, including：
Step S201：Transfer faceform's recognizer.
Step S202：Dividing processing is carried out to the human face scanning model for obtaining according to the faceform's recognizer transferred.
Step S203：Recognize and extract the positive face part of human face scanning model.
In the present embodiment, the human face scanning model for obtaining is divided by transferring default faceform's recognizer
Treatment is cut, the human face scanning model is divided into positive face part and anon-normal face part (such as hair and clothes), then recognized
And positive face part therein is extracted, that is to say characteristic the most typical in manikin.
Step S103：Human face scanning model according to obtaining judges human body sex.
In the present embodiment, typical men and women's feature, such as shape of face, skin can be analyzed by faceform's recognizer
Color, face size etc., the sex of the current human face scanning model of comprehensive descision is man or female.
Step S104：Human body sex according to judging transfers corresponding male/female common model.
Specifically, when the sex for judging current human face scanning model is man, default male common model can be transferred；When
When the sex for judging current human face scanning model is female, default female's common model can be transferred.The common model is basis
The typical standard skull model that the three-dimensional feature statistics of particular country or region man or woman's head is previously generated.
Step S105：The common model transferred of segmentation is extracting the anon-normal face part of common model.
Similar, dividing processing can be carried out by transferring common model of the default faceform's recognizer to transferring,
The common model is divided into positive face part and anon-normal face part (such as hair and clothes), is then recognized and is extracted and be therein
Anon-normal face part, that is to say typical anon-normal face characteristic in common model.
Step S106：The anon-normal face part of the positive face part of the human face scanning model of extraction and common model is sutured
Treatment, to generate face 3D models.
Behind the positive face part for extracting human face scanning model and the anon-normal face part of common model, both are sutured
Process to generate face 3D models.Referring to Fig. 3, the stitching processing generates face 3D models, including：
Step S301：Obtain the size of the anon-normal face part of the positive face part and common model of human face scanning model.
Step S302：The size of the positive face part of the human face scanning model is adjusted, to be adapted to the non-of the common model
The size of positive face part.
Step S303：The positive face part of the human face scanning model after adjustment is entered with the anon-normal face part of the common model
Row stitching processing, to generate face 3D models.
In the present embodiment, the size of the positive face part of adjustment human face scanning model is being adapted to the anon-normal face of common model
Point, stitching processing then is carried out to both and new face 3D models are generated.Newly-generated face 3D models, remain modeler
Typical positive face characteristic in the body scans model of body, while typical anon-normal face features in also having merged common model
Point, relatively low operating efficiency and relatively low standard during compared to mistake in the existing artificial treatment body scans model by art designing or defect
Easily there is mistake and leads in anon-normal face part in exactness, the body scans model that effectively prevent existing 3D body-scanners generation
The error rate and error problem higher of face 3D models are caused, while conveniently and efficiently generating accurate face 3D models, is improved
The success rate of face 3D modeling, accuracy and operating efficiency.
Step S107：Face 3D models to generating are smoothed, and export the face 3D models.
After stitching processing generation face 3D models, it is necessary to positive face part and common model to human face scanning model it is non-
The seam of positive face part optimizes treatment.Specifically, referring to Fig. 4, the face 3D model smoothings treatment, including：
Step S401：Transfer faceform's seam smoothing algorithm.
Step S402：The face 3D models for generating are smoothed according to the faceform's seam smoothing algorithm transferred
To eliminate seam.
After new face 3D models are generated, continue to transfer faceform's seam smoothing algorithm to human face scanning model just
The seam of the anon-normal face part of face part and common model optimizes treatment, it is to avoid sewn seams are caused to face 3D models
Influence and error, improve the accuracy and success rate of face 3D modeling.
In the generation method of the face 3D models of the present embodiment, the positive face part of human face scanning model is extracted in segmentation respectively
With the anon-normal face part of common model, and carry out stitching processing and smoothing processing to both to generate new face 3D models, protect
Typical positive face characteristic in body scans model has been stayed, typical anon-normal face characteristic in common model has been merged, together
When also avoid influence and the error that sewn seams are caused to face 3D models, swept by art designing's artificial treatment human body compared to existing
Operating efficiency and the relatively low degree of accuracy relatively low when mistake in model or defect is retouched, existing 3D body-scanners life is effectively prevent
Into body scans model anon-normal face part easily occur mistake and cause the error rate and error of face 3D models higher
Problem, while conveniently and efficiently generating accurate face 3D models, improves success rate, accuracy and the work of face 3D modeling
Efficiency.
Fig. 5 is the structural representation of the generation system of face 3D models in another embodiment of the present invention.As illustrated,
On the basis of above method embodiment, the generation system 100 of the face 3D models in the present embodiment, including the face being sequentially connected
Scan model acquisition module 10, positive face extraction module 20, Sexual discriminating module 30, common model are transferred module 40, anon-normal face and are carried
Modulus block 50, stitching processing module 60 and smoothing processing and output module 70.
In the present embodiment, when needing to set up face 3D models for specific human body, human body can be placed in the people
Rotated on the such as electric turntable of 3D body-scanners of face scan model acquisition module 10, the human face scanning model acquisition module
The face part of the sensor scan human body of 10 such as described 3D body-scanners, to obtain human face scanning model, the face
Scan model includes the three-dimensional data of face part.
After human face scanning model is obtained by the human face scanning model acquisition module 10 such as 3D body-scanners, institute
State the positive face part that positive face extraction module 20 extracts faceform.Specifically, referring to Fig. 6, the positive face extraction module 20 includes
Faceform's recognizer transfers unit 201, dividing processing unit 202 and identification and extraction unit 203.
Wherein, faceform's recognizer transfers unit 201, for transferring faceform's recognizer；Described point
Processing unit 202 is cut, for transferring faceform's recognizer pair that unit 201 is transferred according to faceform's recognizer
The human face scanning model that the human face scanning model acquisition module 10 is obtained carries out dividing processing；The identification and extraction unit
203, for recognizing and extract the dividing processing of dividing processing unit 202 human face scanning model positive face part.
In the present embodiment, unit 201 is transferred by faceform's recognizer and transfers default faceform's knowledge
Other algorithm carries out dividing processing, the dividing processing to the human face scanning model that the human face scanning model acquisition module 10 is obtained
The human face scanning model is divided into positive face part and anon-normal face part (such as hair and clothes) by unit 202, then described
Recognize and extraction unit 203 is recognized and extracts positive face part therein, that is to say characteristic the most typical in manikin.
The Sexual discriminating module 30, for the human face scanning mould obtained according to the human face scanning model acquisition module 10
Type judges human body sex.Specifically, the Sexual discriminating module 30 can analyze typical man by faceform's recognizer
Female's feature, such as shape of face, the colour of skin, face size etc., the sex of the current human face scanning model of comprehensive descision is man or female.
The common model transfers module 40, and it is right that the human body sex for being judged according to the Sexual discriminating module 30 is transferred
The male/female common model answered.Specifically, when the Sexual discriminating module 30 judges that the sex of current human face scanning model is man
When, the common model transfers module 40 and can transfer default male common model；When the Sexual discriminating module 30 judges currently
The sex of human face scanning model when being female, the common model transfers module 40 and can transfer default female's common model.It is described
Common model is the typical standard previously generated according to the three-dimensional feature of particular country or region man or woman's head statistics
Skull model.
The anon-normal face extraction module 50, common model that module 40 transfers is transferred to carry for splitting the common model
Take the anon-normal face part of common model.Similar, the anon-normal face extraction module 50 can be known by transferring default faceform
Other algorithm transfers the common model that module 40 transfers to the common model and carries out dividing processing, and the common model is divided into
Positive face part and anon-normal face part (such as hair and clothes), then recognize and extract anon-normal face part therein, that is to say public affairs
Typical anon-normal face characteristic in common mode type.
The stitching processing module 60, the positive face of the human face scanning model for the positive face extraction module 20 to be extracted
Dividing the anon-normal face part of the common model extracted with the anon-normal face extraction module 50 carries out stitching processing, to generate face 3D moulds
Type
The positive face part of human face scanning model and the anon-normal face extraction module are extracted in the positive face extraction module 20
Behind the 50 anon-normal face parts for extracting common model, both are carried out stitching processing to generate face by the stitching processing module 60
3D models.Referring to Fig. 7, the stitching processing module 60 includes Size acquiring unit 601, size adjusting unit 602 and suture
Reason unit 603.
Wherein, the Size acquiring unit 601, for obtain human face scanning model positive face part and common model it is non-
The size of positive face part；The size adjusting unit 602, the size of the positive face part for adjusting the human face scanning model,
To be adapted to the size of the anon-normal face part of the common model；The stitching processing unit 603, for by the size adjusting list
The positive face part of human face scanning model after the adjustment of unit 602 carries out stitching processing with the anon-normal face part of the common model, with
Generation face 3D models.
In the present embodiment, the size of the positive face part of the adjustment of the size adjusting unit 602 human face scanning model is with suitable
Anon-normal face part with common model, then described stitching processing unit 603 pairs both carry out stitching processing and generate new people
Face 3D models.The newly-generated face 3D models of the stitching processing unit 603, remain modeling human body body scans model in
Typical positive face characteristic, while typical anon-normal face characteristic in also having merged common model, compared to existing by U.S.
Operating efficiency relatively low during wrong or defect and the relatively low degree of accuracy in work artificial treatment body scans model, effectively prevent existing
There is the body scans model that 3D body-scanners generate mistake easily occur in anon-normal face part and cause the mistake of face 3D models
Rate and error problem higher by mistake, while conveniently and efficiently generate accurate face 3D models, improve face 3D modeling into
Power, accuracy and operating efficiency.
After the stitching processing of stitching processing module 60 generation face 3D models, the smoothing processing and output module 70
Face 3D models to the stitching processing module 60 generation are smoothed and export the face 3D models.Referring to Fig. 8,
The smoothing processing and output module 70 transfer unit 701, the and of smoothing processing unit 702 including faceform's seam smoothing algorithm
Output unit 703.
Wherein, faceform's seam smoothing algorithm transfers unit 701, and for transferring, faceform's seam is smooth to be calculated
Method；The smoothing processing unit 702, for transferring the face that unit 701 is transferred according to faceform's seam smoothing algorithm
Model seam smoothing algorithm is smoothed to eliminate seam to the face 3D models for generating；The output unit 703, is used for
Export the face 3D models after the smoothing processing of smoothing processing unit 702.New people is generated in the stitching processing module 60
After face 3D models, the smoothing processing and output module 70 continue to transfer faceform's seam smoothing algorithm to human face scanning model
Positive face part and the seam of anon-normal face part of common model optimize treatment, it is to avoid sewn seams are to face 3D models
The influence for causing and error, improve the accuracy and success rate of face 3D modeling.
In the generation system 100 of the face 3D models of the present embodiment, positive face extraction module 20 and anon-normal face extraction module
The 50 anon-normal face parts for splitting the positive face part and common model of extracting human face scanning model respectively, stitching processing module 60 pairs two
Person carries out stitching processing and smoothing processing to generate new face 3D models, and typical positive face is special in remaining body scans model
Part is levied, typical anon-normal face characteristic in common model has been merged, while also avoid sewn seams to face 3D models
The influence for causing and error, relatively low work during compared to mistake in the existing artificial treatment body scans model by art designing or defect
Efficiency and the relatively low degree of accuracy, effectively prevent the body scans model of existing 3D body-scanners generation in anon-normal face partial volume
Easily there is mistake and cause the error rate of face 3D models and error problem higher, while conveniently and efficiently generating accurate people
Face 3D models, improve success rate, accuracy and the operating efficiency of face 3D modeling.
It should be appreciated that each several part of the invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In implementation method, the software that multiple steps or method can in memory and by suitable instruction execution system be performed with storage
Or firmware is realized.If for example, realized with hardware, and in another embodiment, can be with well known in the art
Any one of row technology or their combination are realized：With the logic gates for realizing logic function to data-signal
Discrete logic, the application specific integrated circuit with suitable combinational logic gate circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description such as example " or " some examples " means to combine specific features, structure, material or feature that the embodiment or example are described
It is contained at least one embodiment of the invention or example.In this manual, the schematic representation to above-mentioned term differs
Surely identical embodiment or example are referred to.And, the specific features of description, structure, material or feature can be any
Combined in an appropriate manner in one or more embodiments or example.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not
Can these embodiments be carried out with various changes, modification, replacement and modification in the case of departing from principle of the invention and objective, this
The scope of invention is limited by claim and its equivalent.
Claims (10)
1. a kind of generation method of face 3D models, it is characterised in that including：
Obtain human face scanning model；
Split the human face scanning model of acquisition to extract the positive face part of human face scanning model；
Human face scanning model according to obtaining judges human body sex；
Human body sex according to judging transfers corresponding male/female common model；
The common model transferred of segmentation is extracting the anon-normal face part of common model；
The anon-normal face part of the positive face part of the human face scanning model of extraction and common model is carried out into stitching processing, to generate people
Face 3D models；And
Face 3D models to generating are smoothed, and export the face 3D models.
2. the generation method of face 3D models according to claim 1, it is characterised in that the face that the segmentation is obtained is swept
Model is retouched to extract the positive face part of human face scanning model, including：
Transfer faceform's recognizer；
Dividing processing is carried out to the human face scanning model for obtaining according to the faceform's recognizer transferred；And
Recognize and extract the positive face part of human face scanning model.
3. the generation method of face 3D models according to claim 1, it is characterised in that the human face scanning that will be extracted
The positive face part of model and the anon-normal face part of common model carry out stitching processing, to generate face 3D models, including：
Obtain the size of the anon-normal face part of the positive face part and common model of human face scanning model；
The size of the positive face part of the human face scanning model is adjusted, to be adapted to the chi of the anon-normal face part of the common model
It is very little；And
The positive face part of the human face scanning model after adjustment is carried out into stitching processing with the anon-normal face part of the common model, with
Generation face 3D models.
4. the generation method of face 3D models according to claim 1, it is characterised in that the described pair of face 3D mould of generation
Type is smoothed, including：
Transfer faceform's seam smoothing algorithm；And
The face 3D models for generating are smoothed to eliminate seam according to the faceform's seam smoothing algorithm transferred.
5. the generation method of face 3D models according to claim 1, it is characterised in that the acquisition human face scanning mould
Type, including：
Scan face to obtain human face scanning model by 3D body-scanners.
6. a kind of generation system of face 3D models, it is characterised in that including：
Human face scanning model acquisition module, for obtaining human face scanning model；
Positive face extraction module, for splitting the human face scanning model of the human face scanning model acquisition module acquisition to extract face
The positive face part of scan model；
Sexual discriminating module, the human face scanning model for being obtained according to the human face scanning model acquisition module judges human body
Not；
Common model transfers module, and it is public that the human body sex for being judged according to the Sexual discriminating module transfers corresponding male/female
Common mode type；
Anon-normal face extraction module, the common model of module calls is transferred to extract common model for splitting the common model
Anon-normal face part；
Stitching processing module, for the positive face part of human face scanning model for extracting the positive face extraction module and the anon-normal
The anon-normal face part of the common model that face extraction module is extracted carries out stitching processing, to generate face 3D models；And
Smoothing processing and output module, for being smoothed and defeated to the face 3D models that the stitching processing module is generated
Go out the face 3D models.
7. the generation system of face 3D models according to claim 6, it is characterised in that the positive face extraction module bag
Include：
Faceform's recognizer transfers unit, for transferring faceform's recognizer；
Dividing processing unit, for transferring faceform's recognizer pair that unit is transferred according to faceform's recognizer
The human face scanning model that the human face scanning model acquisition module is obtained carries out dividing processing；And
Identification and extraction unit, for recognizing and extract the dividing processing unit dividing processing human face scanning model positive face
Part.
8. the generation system of face 3D models according to claim 6, it is characterised in that the stitching processing module bag
Include：
Size acquiring unit, the size for obtaining the positive face part of human face scanning model and the anon-normal face part of common model；
Size adjusting unit, the size of the positive face part for adjusting the human face scanning model, to be adapted to the common model
Anon-normal face part size；And
Stitching processing unit, for by the size adjusting unit adjust after human face scanning model positive face part and the public affairs
The anon-normal face part of common mode type carries out stitching processing, to generate face 3D models.
9. the generation system of face 3D models according to claim 6, it is characterised in that the smoothing processing and output mould
Block includes：
Faceform's seam smoothing algorithm transfers unit, for transferring faceform's seam smoothing algorithm；
Smoothing processing unit, puts down for transferring faceform's seam that unit transfers according to faceform's seam smoothing algorithm
Sliding algorithm is smoothed to eliminate seam to the face 3D models for generating；And
Output unit, for exporting the face 3D models after the smoothing processing unit smoothing processing.
10. the generation system of face 3D models according to claim 6, it is characterised in that the human face scanning model is obtained
Modulus block is 3D body-scanners.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610998968.7A CN106780714A (en) | 2016-11-10 | 2016-11-10 | The generation method and system of face 3D models |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610998968.7A CN106780714A (en) | 2016-11-10 | 2016-11-10 | The generation method and system of face 3D models |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106780714A true CN106780714A (en) | 2017-05-31 |
Family
ID=58973524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610998968.7A Pending CN106780714A (en) | 2016-11-10 | 2016-11-10 | The generation method and system of face 3D models |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106780714A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110223374A (en) * | 2019-05-05 | 2019-09-10 | 太平洋未来科技（深圳）有限公司 | A kind of pre-set criteria face and head 3D model method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719326A (en) * | 2016-01-19 | 2016-06-29 | 华中师范大学 | Realistic face generating method based on single photo |
| CN205656606U (en) * | 2016-02-26 | 2016-10-19 | 海尔集团技术研发中心 | 3D prints customization image implementation system |
| CN106096580A (en) * | 2016-06-28 | 2016-11-09 | 合肥酷睿网络科技有限公司 | A kind of network by recognition of face method for distinguishing |
-
2016
- 2016-11-10 CN CN201610998968.7A patent/CN106780714A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719326A (en) * | 2016-01-19 | 2016-06-29 | 华中师范大学 | Realistic face generating method based on single photo |
| CN205656606U (en) * | 2016-02-26 | 2016-10-19 | 海尔集团技术研发中心 | 3D prints customization image implementation system |
| CN106096580A (en) * | 2016-06-28 | 2016-11-09 | 合肥酷睿网络科技有限公司 | A kind of network by recognition of face method for distinguishing |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110223374A (en) * | 2019-05-05 | 2019-09-10 | 太平洋未来科技（深圳）有限公司 | A kind of pre-set criteria face and head 3D model method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021238673A1 (en) | Tooth scanning method and apparatus, computer device, and computer-readable storage medium | |
| CN109859203B (en) | Defect tooth image identification method based on deep learning | |
| CN110956635A (en) | Lung segment segmentation method, device, equipment and storage medium | |
| CN107392897B (en) | Organ contour acquisition method, imaging apparatus, radiotherapy planning system, and storage medium | |
| CN111414809B (en) | Three-dimensional pattern recognition method, device, equipment and storage medium | |
| CN109727235B (en) | Organ automatic delineation algorithm based on deep learning | |
| CN108765392B (en) | Digestive tract endoscope lesion detection and identification method based on sliding window | |
| CN106687068A (en) | Method and device for making complete denture based on data mining | |
| CN110097557B (en) | Medical image automatic segmentation method and system based on 3D-UNet | |
| CN110197462A (en) | A kind of facial image beautifies in real time and texture synthesis method | |
| CN111685899A (en) | Dental orthodontic treatment monitoring method based on intraoral images and three-dimensional models | |
| CN110008992B (en) | Deep learning method for prostate cancer auxiliary diagnosis | |
| CN109902682A (en) | A kind of mammary gland x line image detection method based on residual error convolutional neural networks | |
| CN111738989A (en) | Organ delineation method and device | |
| CN112949605A (en) | Semantic segmentation based face makeup method and system | |
| CN110321946A (en) | A kind of Multimodal medical image recognition methods and device based on deep learning | |
| AU2021428496A1 (en) | Artificial intelligence-based method and application for manufacturing 3d prosthesis for tooth restoration | |
| US20220358740A1 (en) | System and Method for Alignment of Volumetric and Surface Scan Images | |
| US20220361992A1 (en) | System and Method for Predicting a Crown and Implant Feature for Dental Implant Planning | |
| CN106709431A (en) | Iris recognition method and device | |
| CN114445409B (en) | Symptom classification has from integrated system for oral diagnosis of contrast function | |
| CN112907581A (en) | MRI (magnetic resonance imaging) multi-class spinal cord tumor segmentation method based on deep learning | |
| CN106780714A (en) | The generation method and system of face 3D models | |
| CN113705466A (en) | Human face facial feature occlusion detection method used for occlusion scene, especially under high-imitation occlusion | |
| CN117745939A (en) | Acquisition method, device, equipment and storage medium of three-dimensional digital model |
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 | ||
| CB02 | Change of applicant information |
Address after: 518000 Guangdong city of Shenzhen province Futian District North Road Home sculpture 18-34a Applicant after: Shenzhen Sanwei Artificial Intelligence Technology Co., Ltd. Address before: 518000 Guangdong city of Shenzhen province Futian District North Road Home sculpture 18-34a Applicant before: Shenzhen coffee restaurant consultant Co., Ltd. |
|
| CB02 | Change of applicant information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Wang Zhiquan Inventor after: Wang Haomian Inventor after: Huang Zhe Inventor before: Wang Zhiquan Inventor before: Liang Jinhua Inventor before: Zeng Jiadong Inventor before: Shen Xiyou Inventor before: Wang Haomian |
|
| CB03 | Change of inventor or designer information | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |
|
| RJ01 | Rejection of invention patent application after publication |