CN108596091A - Figure image cartooning restoring method, system and medium - Google Patents
Figure image cartooning restoring method, system and medium Download PDFInfo
- Publication number
- CN108596091A CN108596091A CN201810373705.6A CN201810373705A CN108596091A CN 108596091 A CN108596091 A CN 108596091A CN 201810373705 A CN201810373705 A CN 201810373705A CN 108596091 A CN108596091 A CN 108596091A
- Authority
- CN
- China
- Prior art keywords
- face
- stylized
- image
- cartoon
- coordinate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of figure image cartooning restoring method, specifically include:Establish cartoon material database；Obtain pending character image information and Skin Color Information；Cartoon style processing is carried out according to character image and Skin Color Information, obtains the position of stylized shape of face, stylized face and stylized face in stylized shape of face；According to five, three front yard of face rule, Expert Rules correction is carried out to accurate location of the stylized face in stylized shape of face, obtains correcting image；Cartoon synthesis is carried out according to correcting image, stylized shape of face and stylized face, obtains cartoon human face image；Build data, hair style data and cartoon human face image are synthesized, cartoon character is obtained.A kind of figure image cartooning restoring method of the present invention, can be according to character image reduction synthesis cartoon character, and the cartoon character of synthesis is both high with the character image similarity of offer, and improves aesthetics.
Description
Technical field
The present invention relates to internet works software technical fields, and in particular to a kind of figure image cartooning restoring method, system
And medium.
Background technology
With the development of face recognition technology, more and more industries start to be combined the technology that face restores with industry,
It is widely used in the industries such as security protection, finance, the administration of justice, public security.Personage's cartoon character designer mostly carries out hand according to personage's picture
Work designs cartoon character or by needing to manually adjust after Software Create, rare at present to be automatically synthesized cartoon character according to character image
System, and cartoon reduction similarity it is not high.
Invention content
For the defects in the prior art, one of the objects of the present invention is to provide a kind of reduction sides of figure image cartooning
Method, can according to character image reduction synthesis cartoon character, the cartoon character of synthesis both with the character image similarity of offer
Height, and improve aesthetics.
The restoring method in a first aspect, a kind of figure image provided in an embodiment of the present invention cartoonizes, specifically includes:It obtains more
Character image classifies to character image according to face feature, physical characteristic and hair style feature as training sample, from dividing
The apparent sample of class extracting data feature carries out cartoon reduction, establishes cartoon material database, the data in the cartoon material database
Including stylized shape of face data, stylized face data, build data and hair style data；
Obtain pending character image information and Skin Color Information；
Carry out cartoon style processing according to character image and Skin Color Information, obtain stylized shape of face, stylized face and
Position of the stylized face in stylized shape of face；
According to five, three front yard of face rule, expert is carried out to accurate location of the stylized face in stylized shape of face
Rule correction, obtains correcting image；
Cartoon synthesis is carried out according to correcting image, stylized shape of face and stylized face, obtains cartoon human face image；
Build data, hair style data and cartoon human face image are synthesized, cartoon character is obtained.
Optionally, the specific method of the cartoon styleization processing includes：Face is carried out to the pending character image
Feature point extraction obtains the location information of face mask characteristic point, face characteristic point and face；
According to face mask characteristic point, face characteristic point and the colour of skin, matched and searched is carried out in cartoon material database, obtains phase
Like the high stylized shape of face of degree and stylized face；
Stylized face are obtained in stylized shape of face according to the location information of face mask characteristic point information and face
Position.
Optionally, according to correcting image, style after obtaining pending character image information and Skin Color Information step
Change shape of face and before stylized face carry out cartoon synthesis step, further include according to facial image, judge on facial image whether
There are glasses.
Optionally, the stylized face include stylized eyebrow, stylized eyes, stylized nose and stylized mouth
Bar.
Optionally, the method for carrying out human face characteristic point extraction to the pending character image specifically includes：According to face
Image establishes coordinate system, and the midpoint of two eye positions is the origin O of coordinate system；The line direction at the center of point O and nose is
The Y-axis of coordinate system；The line direction at the center of point O and eyes is the X-axis of coordinate system；According to the coordinate system of establishment, one is defined
13 feature vector, Xs=(x1,x2,……,x13), wherein：
(x1, x2, x3The x coordinate of)=(left eyebrow position, the y-coordinate of left eyebrow position, the length of eyebrow)；
(x4, x5, x6The x coordinate of)=(left eye position, the y-coordinate of left eye position, the length of left eye eyeball)；
(x7, x8The y-coordinate of)=(nose shape, the width of nose)；
(x9, x10The y-coordinate of)=(face position, the width of face)；
(x11, x12The y-coordinate of)=(cheek position, the width of cheek)；
(x13The y-coordinate of)=(chin portion).
Second aspect, a kind of figure image cartooning provided by the invention go back original system, including cartoon materials database, number
According to acquisition module, stylized processing module, Expert Rules rectification module, cartoon human face image synthesis unit and cartoon character
Synthesis module, wherein
The cartoon materials database is according to multiple character images as training sample, special according to face feature, build
Hair style of seeking peace feature classifies to character image, and the apparent sample of feature is extracted from grouped data and carries out cartoon reduction, is built
Cartoon material database is found, the data in the cartoon material database include stylized shape of face data, stylized face data, build data
With hair style data；
The data acquisition module is for obtaining pending character image information and Skin Color Information；
The stylization processing module is used to carry out cartoon style processing according to character image and Skin Color Information, obtains wind
Position of the shape of face, stylized face and stylized face of formatting in stylized shape of face；
The Expert Rules rectification module is used for according to five, three front yard of face rule, to the stylized face in stylization
Accurate location in shape of face carries out Expert Rules correction, obtains correcting image；
The cartoon human face image synthesis unit is used for according to correcting image, stylization contact and stylized face card
Logical synthesis, obtains cartoon human face image；
The cartoon character synthesis module is for closing body data, hair style data and cartoon human face image
At obtaining cartoon character.
Optionally, the stylized processing module includes feature point extraction module, stylized shape of face processing module, stylization
Face processing module and stylized face position module, the feature point extraction module be used for the pending character image into
Pedestrian's face characteristic point extracts, and obtains the location information of face mask characteristic point, face characteristic point and face；The stylization shape of face
Processing module is used to, according to face mask characteristic point and Skin Color Information, matched and searched is being carried out in cartoon material database, is obtaining phase
Like the high stylized shape of face of degree；The stylization face processing module is used for according to face characteristic point and the colour of skin in cartoon material database
Middle carry out matched and searched obtains the high stylized face of similarity；The stylization face position module according to face for taking turns
The location information of wide characteristic point information and face obtains position of the stylized face in stylized shape of face.
Optionally, the system also includes glasses judgment module, the glasses judgment module is used to, according to facial image, sentence
Whether there are glasses on disconnected facial image.
Optionally, the specific works method of the feature point extraction module includes：Coordinate system is established according to facial image, two
The midpoint of a eye position is the origin O of coordinate system；The line direction at the center of point O and nose is the Y-axis of coordinate system；Point O with
The line direction at the center of eyes is the X-axis of coordinate system；According to the coordinate system of establishment, one 13 feature vector, Xs are defined
=(x1,x2,……,x13), wherein：
(x1, x2, x3The x coordinate of)=(left eyebrow position, the y-coordinate of left eyebrow position, the length of eyebrow)；
(x4, x5, x6The x coordinate of)=(left eye position, the y-coordinate of left eye position, the length of left eye eyeball)；
(x7, x8The y-coordinate of)=(nose shape, the width of nose)；
(x9, x10The y-coordinate of)=(face position, the width of face)；
(x11, x12The y-coordinate of)=(cheek position, the width of cheek)；
(x13The y-coordinate of)=(chin portion).
The third aspect, the embodiment of the present invention also provide a kind of computer readable storage medium, the computer storage media
It is stored with computer program, the computer program includes program instruction, and described program instruction makes institute when being executed by a processor
It states processor and holds the above method.
Beneficial effects of the present invention：
A kind of figure image cartooning restoring method provided in this embodiment, five are carried out according to a large amount of training sample
The features such as official, build, hair style carry out classification of type, corresponding cartoon material database are established, then by pending character image
The identification of feature transfers the highest material of similarity in cartoon material database and carries out reduction synthesis cartoon character.The card of synthesis
Logical figure image was both high with the character image similarity of offer, and improved aesthetics.Pass through a large amount of training samples and material
Amount is continuously increased, and improves the similarity of cartoon figure's reduction, has stronger scalability.
A kind of figure image cartooning provided in this embodiment goes back original system, and five are carried out according to a large amount of training sample
The features such as official, build, hair style carry out classification of type, corresponding cartoon material database are established, then by pending character image
The identification of feature transfers the highest material of similarity in cartoon material database and carries out reduction synthesis cartoon character.The card of synthesis
Logical figure image was both high with the character image similarity of offer, and improved aesthetics.Pass through a large amount of training samples and material
Amount is continuously increased, and improves the similarity of cartoon figure's reduction, has stronger scalability.
Description of the drawings
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art are briefly described.In all the appended drawings, similar element
Or part is generally identified by similar reference numeral.In attached drawing, each element or part might not be drawn according to actual ratio.
Fig. 1 shows a kind of flow chart of figure image cartooning restoring method first embodiment provided by the present invention；
Fig. 2 shows the particular flow sheets of step S103 in Fig. 1；
Fig. 3 shows that a kind of figure image cartooning provided by the present invention goes back the principle frame of original system first embodiment
Figure；
Fig. 4 shows a kind of structural frames of figure image cartooning provided by the present invention also former terminal first embodiment
Figure.
Specific implementation mode
The embodiment of technical solution of the present invention is described in detail below in conjunction with attached drawing.Following embodiment is only used for
Clearly illustrate technical scheme of the present invention, therefore be only used as example, and the protection model of the present invention cannot be limited with this
It encloses.It should be noted that unless otherwise indicated, technical term or scientific terminology used in this application are should be belonging to the present invention
The ordinary meaning that field technology personnel are understood.
It should be appreciated that ought use in this specification and in the appended claims, term " comprising " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, element, component and/or its presence or addition gathered.
It is also understood that the term used in this description of the invention is merely for the sake of the mesh for describing specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singulative, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combinations and all possible combinations of one or more of associated item listed, and includes these combinations.
As used in this specification and in the appended claims, term " if " can be according to context quilt
Be construed to " when ... " or " once " or " in response to determination " or " in response to detecting ".
As shown in Figure 1, Fig. 1 shows a kind of flow of figure image cartooning restoring method provided in an embodiment of the present invention
Figure, this method specifically include following steps：
S101：Multiple character images are obtained as training sample, according to face feature, physical characteristic and hair style feature to people
Object image is classified, and the apparent sample of feature is extracted from grouped data and carries out cartoon reduction, establishes cartoon material database, described
Data in cartoon material database include stylized shape of face data, stylized face data, build data and hair style data.
A large amount of true man's character image sample is collected as training sample, according to face feature, physical characteristic and hair style spy
Sign classifies to character image.Face is divided by 5 shape of face, eyebrow, eyes, nose, lip dimensions.Shape of face is drawn
It is divided into 5 kinds of round face, rectangular face, heart-shaped face, oval face and diamond shape face；By eyebrow be divided into nature eyebrow, straight eyebrow, arched eyebrows,
Arch eyebrow and above choose 5 type of eyebrow；Nose is divided into 3 type of standard nose, curve nose and straight line nose；Eyes are divided into standard
Eye, the eye that hangs down, slim eye, birdeye, hypertropia, elongated eye, circle eye and 8 type of single-edge eyelid；Lip is divided into standard lip, thin
Lip, thick lips, the corners of the mouth upwarp and 5 type of upper-thin-lower-thick, total 26 types.By taking the eyes in face as an example, first find
3000 samples, classify according to the type of face, then compare eye contour and delete the similar eyes of feature, until surplus
Lower 5 most typical samples.Other parts and so on, 26 class face parts delete per class and select 5 typical samples, amount to 130
A local feature.By build, age-based, 4 build, the colour of skin, garment language dimensions divide, then will be divided into 18 at the age
Year old or less, 18-25 Sui, 26-30 Sui, 31-35 Sui, 6 classes of 36-40 Sui and 40 years old or more, by build be divided into it is slender, always, it is rich
Fertile 3 classes；The colour of skin is divided into pale, intermediate and healthy 3 class；Garment language is divided into sweet and lovely, naturally graceful, romantic fan
People, glamour fashion, graceful soft and graceful, magnificent graceful, pure succinct, intellectual calmness, modern 9 kinds of modern times, each type is multiplied to obtain
486 kinds of physical characteristic combinations.Hair style is divided by 5 length, bang, profile, texture, color development dimensions, then will be sent out dash
Be divided into long hair, middle length, BOB, bob, ultrashort 5 class, during bang is divided into point, side point, short bang, arc bang, neat bang,
Air bang, wide bang, in divide bang, oblique bang, 10 class of high bang；Profile is divided into high center of gravity, middle center of gravity, centre of gravity 3
Class；Texture is divided into straight hair, micro- volume, half volume, big volume, 5 class of rouleau；It is cold that color development is divided into high brightness cool colour, middle lightness
Colour system, low lightness cool colour, high brightness neutrality colour system, middle lightness neutrality colour system, low lightness neutrality colour system, high brightness warm colour,
Each type is multiplied to obtain 6750 kinds of hair style combinations by middle lightness warm colour, low 9 class of lightness warm colour.From these grouped datas
In extract the apparent sample of feature again and carry out cartoon reduction, establish cartoon material database, the data in cartoon material database include style
Change shape of face data, stylized face data, build data and hair style data.
S102：Obtain pending character image information and Skin Color Information.It can be inputted in specific implementation by user
Pending character image information and Skin Color Information can also pass through the pending character image information of image acquisition device
Or other modes obtain character image, judge Skin Color Information according to character image information.It uses in the present embodiment and passes through user
Input the mode of pending character image information and Skin Color Information.
S103：Cartoon style processing is carried out according to pending character image and Skin Color Information, obtains stylized shape of face, wind
It formats the position of face and stylized face in stylized shape of face.Stylized face include stylized eyebrow, stylized eyes,
Stylized nose and stylized face.
The detection of personage's style is using style Aesthetic Theory current on the market, the basic judgement of personage's style Aesthetic Theory：
First, sort out to the profile attributes of each section of people, according to curvature difference, can substantially be divided into it is bent, in, straight three classes；Second is that
The amount sense attribute of each section of people is sorted out, according to area difference, be broadly divided into it is soft, in, hard three classes, then will be above-mentioned
Classification results carry out assembled classification again in the way of meeting human cognitive, can obtain nine different combinations, referred to as " nine types
The rationality conclusion to style aesthetic feeling is consequently formed in style ".Such as：Profile is bent, amount feels small combination, and often embodying dexterity can
The feature of love, so being classified as " sweet and lovely " style；And profile is straight, amount feels big combination, then embodies strong, Modern Beauty
Feature is then concluded as " modern modern " style.Main basis " nine type styles " is to pending image pattern in the present embodiment
Shape of face and face carry out stylized processing.Because face and face are the significant points of people's overall image, face and face
Whether whether true, exact effect entirety cartoon character is true, accurate for cartoon styleization processing.
Specifically, as shown in Fig. 2, carrying out the specific of cartoon style processing according to pending character image and Skin Color Information
Method includes：S1031：Human face characteristic point extraction is carried out to the pending character image, obtains face mask characteristic point, face
The location information of characteristic point and face.The method for carrying out human face characteristic point extraction to pending character image specifically includes：According to
Facial image establishes coordinate system, sets the midpoint of two eye positions as the origin O of coordinate system；The company at the center of point O and nose
Line direction is the Y-axis of coordinate system；The line direction at the center of point O and eyes is the X-axis of coordinate system；According to the coordinate system of establishment,
Define one 13 feature vector, X=(x1,x2,……,x13), wherein：
(x1, x2, x3The x coordinate of)=(left eyebrow position, the y-coordinate of left eyebrow position, the length of eyebrow)；
(x4, x5, x6The x coordinate of)=(left eye position, the y-coordinate of left eye position, the length of left eye eyeball)；
(x7, x8The y-coordinate of)=(nose shape, the width of nose)；
(x9, x10The y-coordinate of)=(face position, the width of face)；
(x11, x12The y-coordinate of)=(cheek position, the width of cheek)；
(x13The y-coordinate of)=(chin portion), the lowest point of the y coordinate representation human face characteristic point of chin portion.
S1032：According to face mask characteristic point, face characteristic point and colour of skin image, matched in cartoon material database
It searches, obtains the high stylized shape of face of similarity and stylized face.
S1033：Stylized face are obtained in stylized face according to the location information of face mask characteristic point information and face
Position in type.According to the extraction of human face characteristic point, length, width and the coordinate position of face are obtained, in conjunction with face mask spy
The position of sign point and face obtains accurate location of the stylized face in stylized shape of face.
Method further includes S107：According to facial image, judge whether there are glasses on facial image, if there is glasses, synthesis
Glasses information.Whether there is the specific method identification of glasses that the prior art or existing interface realization can be used on facial image.Pass through
To whether thering are glasses to judge on facial image, it can more meet character image when later stage cartoon synthesizes, make cartoon character more
It is true to nature.
S104：According to five, three front yard of face rule, to accurate location of the stylized face in stylized shape of face into
Row Expert Rules are corrected, and correcting image is obtained.The position of face is corrected according to three front yard of face, five ratios so that five
The characteristics of there is preferable balance in official position, can not only highlight personage itself between fitting reality and beauty, but also improve beauty
Degree.
S105：Cartoon synthesis is carried out according to correcting image, stylized shape of face and stylized face, obtains cartoon human face figure
Picture.If it is determined that character image there are glasses, blocked according to correcting image, stylized shape of face, stylized face and glasses information
Logical synthesis, obtains cartoon human face image.
S106：Build data, hair style data and cartoon human face image are synthesized, cartoon character is obtained.By people
Height, weight, bang, hairline and the cartoon human face image of object synthesize cartoon character.The cartoon character of synthesis
Both high with the character image similarity of offer, and improve aesthetics.Pass through the continuous increasing of a large amount of training samples and material amount
Add, improves the similarity of cartoon figure's reduction, have stronger scalability.
Figure image cartooning restoring method can be applied in hair style design, can be high with the hair style perfect adaptation of reduction
The relationship of degree reduction personage and hair style select to provide reference when hair style for consumer, have great commercial value.
A kind of figure image cartooning restoring method provided in this embodiment, five are carried out according to a large amount of training sample
The features such as official, build, hair style carry out classification of type, corresponding cartoon material database are established, then by pending character image
The identification of feature transfers the highest material of similarity in cartoon material database and carries out reduction synthesis cartoon character.The card of synthesis
Logical figure image was both high with the character image similarity of offer, and improved aesthetics.Pass through a large amount of training samples and material
Amount is continuously increased, and improves the similarity of cartoon figure's reduction, has stronger scalability.
As shown in figure 3, a kind of figure image cartooning provided in an embodiment of the present invention goes back the functional block diagram of original system, this is
System includes cartoon materials database 201, data acquisition module 202, stylized processing module 203, Expert Rules rectification module
204, cartoon human face image synthesis unit 205 and cartoon character synthesis module 206, wherein
The cartoon materials database 201 is according to multiple character images as training sample, according to face feature, build
Feature and hair style feature classify to character image, and the apparent sample of feature is extracted from grouped data and carries out cartoon reduction,
Cartoon material database is established, the data in the cartoon material database include stylized shape of face data, stylized face data, build number
According to hair style data.
The data acquisition module 202 is for obtaining pending character image information and Skin Color Information.
The stylization processing module 203 is used to carry out cartoon style processing according to character image and Skin Color Information, obtains
To the position of stylized shape of face, stylized face and stylized face in stylized shape of face.Stylized processing module includes spy
Sign point extraction module, stylized shape of face processing module, stylized face processing module and stylized face position module, the spy
Sign point extraction module is used to carry out human face characteristic point extraction to the pending character image, obtains face mask characteristic point, five
The location information of official's characteristic point and face；The stylization shape of face processing module is used to be believed according to face mask characteristic point and the colour of skin
Breath, matched and searched is being carried out in cartoon material database, is obtaining the high stylized shape of face of similarity；The stylization face handle mould
Block is used to carry out matched and searched in cartoon material database according to face characteristic point and the colour of skin, obtains the high stylization five of similarity
Official；The stylization face position module is used to obtain stylization according to the location information of face mask characteristic point information and face
Position of the face in stylized shape of face.The specific works method of feature point extraction module includes：It is established and is sat according to facial image
Mark system, the midpoint of two eye positions are the origin O of coordinate system；The line direction at the center of point O and nose is the Y of coordinate system
Axis；The line direction at the center of point O and eyes is the X-axis of coordinate system；According to the coordinate system of establishment, one 13 spies are defined
Levy vector X=(x1,x2,……,x13), wherein：
(x1, x2, x3The x coordinate of)=(left eyebrow position, the y-coordinate of left eyebrow position, the length of eyebrow)；
(x4, x5, x6The x coordinate of)=(left eye position, the y-coordinate of left eye position, the length of left eye eyeball)；
(x7, x8The y-coordinate of)=(nose shape, the width of nose)；
(x9, x10The y-coordinate of)=(face position, the width of face)；
(x11, x12The y-coordinate of)=(cheek position, the width of cheek)；
(x13The y-coordinate of)=(chin portion).
The Expert Rules rectification module 204 is used for according to five, three front yard of face rule, to the stylized face in wind
Accurate location in shape of face of formatting carries out Expert Rules correction, obtains correcting image.According to the ratio in five, three front yard of face by five
It corrects the position of official so that face position has preferable balance between fitting reality and beauty, can both highlight personage
Itself the characteristics of, and improve aesthetics.
The cartoon human face image synthesis unit 205 be used for according to correcting image, stylization contact and stylized face into
Row cartoon synthesizes, and obtains cartoon human face image.
The cartoon character synthesis module 206 is used to carry out body data, hair style data and cartoon human face image
Synthesis, obtains cartoon character.The height of personage, weight, bang, hairline and cartoon human face image are synthesized into cartoon character
Figure image.The cartoon character of synthesis was both high with the character image similarity of offer, and improved aesthetics.By a large amount of
Training sample and material amount are continuously increased, and improve the similarity of cartoon figure's reduction, have stronger scalability.
The system further includes glasses judgment module 207, and the glasses judgment module 207 is used to, according to facial image, judge
Whether there are glasses on facial image.If it is determined that character image has glasses, according to correcting image, stylized shape of face, stylization five
Official and glasses information carry out cartoon synthesis, obtain cartoon human face image.
A kind of figure image cartooning provided in this embodiment goes back original system, and five are carried out according to a large amount of training sample
The features such as official, build, hair style carry out classification of type, corresponding cartoon material database are established, then by pending character image
The identification of feature transfers the highest material of similarity in cartoon material database and carries out reduction synthesis cartoon character.The card of synthesis
Logical figure image was both high with the character image similarity of offer, and improved aesthetics.Pass through a large amount of training samples and material
Amount is continuously increased, and improves the similarity of cartoon figure's reduction, has stronger scalability.
The embodiment of former terminal is gone back cartoonizing the present invention also provides a kind of figure image, as shown in figure 4, the terminal includes
Processor 301, input equipment 302, output equipment 303 and memory 304, the processor 301, input equipment 302, output are set
It is connected with each other for 303 and memory 304, for the memory 304 for storing computer program, the computer program includes journey
Sequence instructs, and the processor 301 is configured for calling described program instruction, executes the above method.
It should be appreciated that in embodiments of the present invention, alleged processor 301 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), application-specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..General processor can be microprocessor or this at
It can also be any conventional processor etc. to manage device.
Input equipment 302 may include that Trackpad, fingerprint adopt sensor (finger print information and fingerprint for acquiring user
Directional information), microphone etc., output equipment 803 may include display (LCD etc.), loud speaker etc..
The memory 304 may include read-only memory and random access memory, and to processor 801 provide instruction and
Data.The a part of of memory 304 can also include nonvolatile RAM.For example, memory 304 can also be deposited
Store up the information of device type.
In the specific implementation, processor 301 described in the embodiment of the present invention, input equipment 302, output equipment 303 can
Realization method described in embodiment of the method provided in an embodiment of the present invention is executed, also can perform described in the embodiment of the present invention
The realization method of system, details are not described herein.
The present invention also provides a kind of computer readable storage medium embodiment, the computer storage media is stored with meter
Calculation machine program, the computer program include program instruction, and described program instruction makes the processor when being executed by a processor
The method for holding above-described embodiment description.
Computer readable storage medium, the computer storage media are stored with computer program, the computer program
Including program instruction, described program instruction makes the processor execute the side that above-described embodiment describes when being executed by a processor
Method.
The computer readable storage medium can be the internal storage unit of the terminal described in previous embodiment, such as eventually
The hard disk or memory at end.The computer readable storage medium can also be the External memory equipment of the terminal, such as described
The plug-in type hard disk being equipped in terminal, intelligent memory card (Smart Media Card, SMC), secure digital (Secure
Digital, SD) card, flash card (Flash Card) etc..Further, the computer readable storage medium can also be wrapped both
The internal storage unit for including the terminal also includes External memory equipment.The computer readable storage medium is described for storing
Other programs needed for computer program and the terminal and data.The computer readable storage medium can be also used for temporarily
When store the data that has exported or will export.
Those of ordinary skill in the art may realize that lists described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, depend on the specific application and design constraint of technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
It is apparent to those skilled in the art that for convenience of description and succinctly, the end of foregoing description
The specific work process at end and unit, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed terminal and method, it can be by other
Mode realize.For example, the apparatus embodiments described above are merely exemplary, for example, the division of the unit, only
For a kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or component can combine
Or it is desirably integrated into another system, or some features can be ignored or not executed.In addition, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some interfaces, device or unit
It connects, can also be electricity, mechanical or other form connections.
Finally it should be noted that：The above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Present invention has been described in detail with reference to the aforementioned embodiments for pipe, it will be understood by those of ordinary skill in the art that：Its according to
So can with technical scheme described in the above embodiments is modified, either to which part or all technical features into
Row equivalent replacement；And these modifications or replacements, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover in the claim of the present invention and the range of specification.
Claims (10)
- The restoring method 1. a kind of figure image cartoonizes, which is characterized in that specifically include following steps：Multiple character images are obtained as training sample, according to face feature, physical characteristic and hair style feature to character image into Row classification extracts the apparent sample of feature from grouped data and carries out cartoon reduction, establishes cartoon material database, the cartoon material Data in library include stylized shape of face data, stylized face data, build data and hair style data；Obtain pending character image information and Skin Color Information；Cartoon style processing is carried out according to character image and Skin Color Information, obtains stylized shape of face, stylized face and style Change position of the face in stylized shape of face；According to five, three front yard of face rule, Expert Rules are carried out to accurate location of the stylized face in stylized shape of face Correction, obtains correcting image；Cartoon synthesis is carried out according to correcting image, stylized shape of face and stylized face, obtains cartoon human face image；Build data, hair style data and cartoon human face image are synthesized, cartoon character is obtained.
- The restoring method 2. figure image as described in claim 1 cartoonizes, which is characterized in that the cartoon styleization processing Specific method includes：Human face characteristic point extraction is carried out to the pending character image, it is special to obtain face mask characteristic point, face The location information of sign point and face；According to face mask characteristic point, face characteristic point and the colour of skin, matched and searched is carried out in cartoon material database, obtains similarity High stylized shape of face and stylized face；Position of the stylized face in stylized shape of face is obtained according to the location information of face mask characteristic point information and face.
- The restoring method 3. figure image as described in claim 1 cartoonizes, which is characterized in that obtaining pending figure map As after information and Skin Color Information step according to correcting image, stylized shape of face and stylized face carry out cartoon synthesis step it Before, further include：According to facial image, judge whether there are glasses on facial image.
- The restoring method 4. figure image as described in claim 1 cartoonizes, which is characterized in that the stylization face include wind It formats eyebrow, stylized eyes, stylized nose and stylized face.
- The restoring method 5. figure image as claimed in claim 2 cartoonizes, which is characterized in that the pending character image The method for carrying out human face characteristic point extraction specifically includes：Coordinate system is established according to facial image, the midpoint of two eye positions is The origin O of coordinate system；The line direction at the center of point O and nose is the Y-axis of coordinate system；The line side at the center of point O and eyes To the X-axis for coordinate system；According to the coordinate system of establishment, one 13 feature vector, X=(x are defined1,x2,……,x13), Wherein：(x1, x2, x3The x coordinate of)=(left eyebrow position, the y-coordinate of left eyebrow position, the length of eyebrow)；(x4, x5, x6The x coordinate of)=(left eye position, the y-coordinate of left eye position, the length of left eye eyeball)；(x7, x8The y-coordinate of)=(nose shape, the width of nose)；(x9, x10The y-coordinate of)=(face position, the width of face)；(x11, x12The y-coordinate of)=(cheek position, the width of cheek)；(x13The y-coordinate of)=(chin portion).
- 6. a kind of figure image cartooning goes back original system, which is characterized in that including cartoon materials database, data acquisition module, Stylized processing module, Expert Rules rectification module, cartoon human face image synthesis unit and cartoon character synthesis module, In,The cartoon materials database is according to multiple character images as training sample, according to face feature, physical characteristic and Hair style feature classifies to character image, and the apparent sample of feature is extracted from grouped data and carries out cartoon reduction, establishes card Lead to material database, the data in the cartoon material database include stylized shape of face data, stylized face data, build data and hair Type data；The data acquisition module is for obtaining pending character image information and Skin Color Information；The stylization processing module is used to carry out cartoon style processing according to character image and Skin Color Information, obtains stylization The position of shape of face, stylized face and stylized face in stylized shape of face；The Expert Rules rectification module is used for according to five, three front yard of face rule, to the stylized face in stylized shape of face In accurate location carry out Expert Rules correction, obtain correcting image；The cartoon human face image synthesis unit is used to carry out cartoon conjunction according to correcting image, stylization contact and stylized face At obtaining cartoon human face image；The cartoon character synthesis module is obtained for synthesizing body data, hair style data and cartoon human face image To cartoon character.
- 7. figure image cartooning as claimed in claim 6 goes back original system, which is characterized in that the stylization processing module packet Include feature point extraction module, stylized shape of face processing module, stylized face processing module and stylized face position module, institute Feature point extraction module is stated for carrying out human face characteristic point extraction to the pending character image, obtains face mask feature The location information of point, face characteristic point and face；The stylization shape of face processing module be used for according to face mask characteristic point and Skin Color Information is carrying out matched and searched in cartoon material database, obtains the high stylized shape of face of similarity；The stylization face Processing module is used to carry out matched and searched in cartoon material database according to face characteristic point and the colour of skin, obtains the high style of similarity Change face；The stylization face position module is used to obtain wind according to the location information of face mask characteristic point information and face It formats position of the face in stylized shape of face.
- 8. figure image cartooning as claimed in claim 6 goes back original system, which is characterized in that the system also includes glasses to sentence Disconnected module, the glasses judgment module are used to, according to facial image, judge whether there is glasses on facial image.
- 9. figure image cartooning as claimed in claim 6 goes back original system, which is characterized in that the feature point extraction module Specific works method includes：Coordinate system is established according to facial image, the midpoint of two eye positions is the origin O of coordinate system；Point O Line direction with the center of nose is the Y-axis of coordinate system；The line direction at the center of point O and eyes is the X-axis of coordinate system；Root According to the coordinate system of establishment, one 13 feature vector, X=(x are defined1,x2,……,x13), wherein：(x1, x2, x3The x coordinate of)=(left eyebrow position, the y-coordinate of left eyebrow position, the length of eyebrow)；(x4, x5, x6The x coordinate of)=(left eye position, the y-coordinate of left eye position, the length of left eye eyeball)；(x7, x8The y-coordinate of)=(nose shape, the width of nose)；(x9, x10The y-coordinate of)=(face position, the width of face)；(x11, x12The y-coordinate of)=(cheek position, the width of cheek)；(x13The y-coordinate of)=(chin portion).
- 10. a kind of computer readable storage medium, which is characterized in that the computer storage media is stored with computer program, The computer program includes program instruction, and described program instruction makes the processor execute such as right when being executed by a processor It is required that 1-5 any one of them methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810373705.6A CN108596091A (en) | 2018-04-24 | 2018-04-24 | Figure image cartooning restoring method, system and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810373705.6A CN108596091A (en) | 2018-04-24 | 2018-04-24 | Figure image cartooning restoring method, system and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108596091A true CN108596091A (en) | 2018-09-28 |
Family
ID=63614468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810373705.6A Pending CN108596091A (en) | 2018-04-24 | 2018-04-24 | Figure image cartooning restoring method, system and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108596091A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109308682A (en) * | 2018-10-18 | 2019-02-05 | 杭州数为科技有限公司 | A kind of recognition of face and cartoon restoring method, device, medium and system |
| CN109408940A (en) * | 2018-10-18 | 2019-03-01 | 杭州数为科技有限公司 | A kind of identification of hair style and restoring method, apparatus and system |
| CN109461118A (en) * | 2018-11-12 | 2019-03-12 | 泰普智能有限公司 | A kind of image processing method and device |
| CN109508708A (en) * | 2019-01-15 | 2019-03-22 | 广州唯品会研究院有限公司 | Measurement method, device and the computer readable storage medium of body data |
| CN109584145A (en) * | 2018-10-15 | 2019-04-05 | 深圳市商汤科技有限公司 | Cartoonize method and apparatus, electronic equipment and computer storage medium |
| CN110503700A (en) * | 2019-08-26 | 2019-11-26 | 北京达佳互联信息技术有限公司 | Generate method, apparatus, electronic equipment and the storage medium of virtual expression |
| CN110648384A (en) * | 2019-06-19 | 2020-01-03 | 北京巴别时代科技股份有限公司 | Cartoon stylized rendering method |
| CN111260763A (en) * | 2020-01-21 | 2020-06-09 | 厦门美图之家科技有限公司 | Cartoon image generation method, device, equipment and storage medium based on portrait |
| CN111460910A (en) * | 2020-03-11 | 2020-07-28 | 深圳市新镜介网络有限公司 | Face type classification method and device, terminal equipment and storage medium |
| CN111667400A (en) * | 2020-05-30 | 2020-09-15 | 温州大学大数据与信息技术研究院 | Human face contour feature stylization generation method based on unsupervised learning |
| CN111882389A (en) * | 2020-07-23 | 2020-11-03 | 上海鸿臣互动传媒有限公司 | Customization system applied to secondary processing of materials |
| CN112164135A (en) * | 2020-09-30 | 2021-01-01 | 山西智优利民健康管理咨询有限公司 | Virtual character image construction device and method |
| CN112926554A (en) * | 2021-04-27 | 2021-06-08 | 南京甄视智能科技有限公司 | Construction of training data set of portrait cartoon stylized model and model generation |
| CN113191847A (en) * | 2021-05-13 | 2021-07-30 | 山东圣梵尼服饰股份有限公司 | Advanced garment customization system |
| CN113409454A (en) * | 2021-07-14 | 2021-09-17 | 北京百度网讯科技有限公司 | Face image processing method and device, electronic equipment and storage medium |
| US11195308B2 (en) | 2018-12-05 | 2021-12-07 | Sony Group Corporation | Patcher tool |
| CN117392330A (en) * | 2023-12-11 | 2024-01-12 | 江西省映尚科技有限公司 | Method and system for manufacturing metauniverse virtual digital person |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101847268A (en) * | 2010-04-29 | 2010-09-29 | 北京中星微电子有限公司 | Cartoon human face image generation method and device based on human face images |
| CN103456010A (en) * | 2013-09-02 | 2013-12-18 | 电子科技大学 | Human face cartoon generation method based on feature point localization |
| CN104637035A (en) * | 2015-02-15 | 2015-05-20 | 百度在线网络技术（北京）有限公司 | Method, device and system for generating cartoon face picture |
| CN104933742A (en) * | 2015-05-25 | 2015-09-23 | 中山大学 | Automatic cartoon image generation method |
| CN105374055A (en) * | 2014-08-20 | 2016-03-02 | 腾讯科技（深圳）有限公司 | Image processing method and device |
-
2018
- 2018-04-24 CN CN201810373705.6A patent/CN108596091A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101847268A (en) * | 2010-04-29 | 2010-09-29 | 北京中星微电子有限公司 | Cartoon human face image generation method and device based on human face images |
| CN103456010A (en) * | 2013-09-02 | 2013-12-18 | 电子科技大学 | Human face cartoon generation method based on feature point localization |
| CN105374055A (en) * | 2014-08-20 | 2016-03-02 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN104637035A (en) * | 2015-02-15 | 2015-05-20 | 百度在线网络技术（北京）有限公司 | Method, device and system for generating cartoon face picture |
| CN104933742A (en) * | 2015-05-25 | 2015-09-23 | 中山大学 | Automatic cartoon image generation method |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109584145A (en) * | 2018-10-15 | 2019-04-05 | 深圳市商汤科技有限公司 | Cartoonize method and apparatus, electronic equipment and computer storage medium |
| CN109408940A (en) * | 2018-10-18 | 2019-03-01 | 杭州数为科技有限公司 | A kind of identification of hair style and restoring method, apparatus and system |
| CN109308682A (en) * | 2018-10-18 | 2019-02-05 | 杭州数为科技有限公司 | A kind of recognition of face and cartoon restoring method, device, medium and system |
| CN109461118A (en) * | 2018-11-12 | 2019-03-12 | 泰普智能有限公司 | A kind of image processing method and device |
| US11195308B2 (en) | 2018-12-05 | 2021-12-07 | Sony Group Corporation | Patcher tool |
| CN109508708A (en) * | 2019-01-15 | 2019-03-22 | 广州唯品会研究院有限公司 | Measurement method, device and the computer readable storage medium of body data |
| CN110648384A (en) * | 2019-06-19 | 2020-01-03 | 北京巴别时代科技股份有限公司 | Cartoon stylized rendering method |
| CN110648384B (en) * | 2019-06-19 | 2023-01-03 | 北京巴别时代科技股份有限公司 | Cartoon stylized rendering method |
| CN110503700A (en) * | 2019-08-26 | 2019-11-26 | 北京达佳互联信息技术有限公司 | Generate method, apparatus, electronic equipment and the storage medium of virtual expression |
| CN111260763A (en) * | 2020-01-21 | 2020-06-09 | 厦门美图之家科技有限公司 | Cartoon image generation method, device, equipment and storage medium based on portrait |
| CN111460910A (en) * | 2020-03-11 | 2020-07-28 | 深圳市新镜介网络有限公司 | Face type classification method and device, terminal equipment and storage medium |
| CN111667400A (en) * | 2020-05-30 | 2020-09-15 | 温州大学大数据与信息技术研究院 | Human face contour feature stylization generation method based on unsupervised learning |
| CN111882389A (en) * | 2020-07-23 | 2020-11-03 | 上海鸿臣互动传媒有限公司 | Customization system applied to secondary processing of materials |
| CN111882389B (en) * | 2020-07-23 | 2024-01-23 | 上海观池聚光数字科技有限公司 | Customization system applied to secondary processing of materials |
| CN112164135A (en) * | 2020-09-30 | 2021-01-01 | 山西智优利民健康管理咨询有限公司 | Virtual character image construction device and method |
| CN112926554A (en) * | 2021-04-27 | 2021-06-08 | 南京甄视智能科技有限公司 | Construction of training data set of portrait cartoon stylized model and model generation |
| CN112926554B (en) * | 2021-04-27 | 2022-08-16 | 南京甄视智能科技有限公司 | Construction of training data set of portrait cartoon stylized model and model generation |
| CN113191847A (en) * | 2021-05-13 | 2021-07-30 | 山东圣梵尼服饰股份有限公司 | Advanced garment customization system |
| CN113409454A (en) * | 2021-07-14 | 2021-09-17 | 北京百度网讯科技有限公司 | Face image processing method and device, electronic equipment and storage medium |
| CN117392330A (en) * | 2023-12-11 | 2024-01-12 | 江西省映尚科技有限公司 | Method and system for manufacturing metauniverse virtual digital person |
| CN117392330B (en) * | 2023-12-11 | 2024-03-08 | 江西省映尚科技有限公司 | Method and system for manufacturing metauniverse virtual digital person |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108596091A (en) | Figure image cartooning restoring method, system and medium | |
| KR101259662B1 (en) | Face classifying method face classifying device classification map face classifying program recording medium where this program is recorded | |
| CN107169455A (en) | Face character recognition methods based on depth local feature | |
| CN106709781A (en) | Personal image design and collocation purchasing device and method | |
| CN109063560A (en) | Image processing method, device, computer readable storage medium and terminal | |
| US8997757B1 (en) | Golden ratio eyebrow shaping method | |
| CN109310196A (en) | Makeup auxiliary device and cosmetic auxiliary method | |
| CN105354869B (en) | Method and system for representing real head features of user on virtual head portrait | |
| CN111860250B (en) | Image recognition method and device based on fine-grained character features | |
| KR20190024582A (en) | Display method for recommending eyebrow style and electronic apparatus thereof | |
| CN110866139A (en) | Cosmetic treatment method, device and equipment | |
| KR20150064977A (en) | Video analysis and visualization system based on face information | |
| CN110728242A (en) | Image matching method and device based on portrait recognition, storage medium and application | |
| CN110377201A (en) | Terminal equipment control method, device, computer installation and readable storage medium storing program for executing | |
| CN106446207B (en) | Makeups library banking process, personalized makeups householder method and its device | |
| CN109727071A (en) | Method and system for advertisement recommendation | |
| CN110490711A (en) | A kind of personalized visualization design platform | |
| CN112906585B (en) | Intelligent hairdressing auxiliary system, method and readable medium based on machine learning | |
| Liu et al. | Magic mirror: An intelligent fashion recommendation system | |
| JP2004145625A (en) | Device for preparing portrait | |
| CN110458751A (en) | A kind of face replacement method, equipment and medium based on Guangdong opera picture | |
| CN106445107A (en) | Method and device for sending information on basis of facial expressions | |
| JP7415387B2 (en) | Virtual character generation device and program | |
| JP2003016431A (en) | Generating device for portrait | |
| CN108985890A (en) | A kind of hair style matching process and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180928 |