CN106204698A

CN106204698A - Virtual image for independent assortment creation generates and uses the method and system of expression

Info

Publication number: CN106204698A
Application number: CN201510227900.4A
Authority: CN
Inventors: 陈容海; 刘永健; 张以纬
Original assignee: Beijing Blue Rhino Space Technology Ltd
Current assignee: Beijing Blue Rhino Space Technology Ltd
Priority date: 2015-05-06
Filing date: 2015-05-06
Publication date: 2016-12-07
Also published as: WO2016177290A1

Abstract

The present invention relates to a kind of virtual image for independent assortment creation generate and use the implementation method of expression, including: select face's organ or local face as parts basis of formation virtual image from a virtual image organ set；Select particular emotion；Then substitute the organ parts in described basic avatars with qualified organ parts, make virtual image meet selected particular emotion.Present invention also offers a kind of generation method in virtual image organ data storehouse, meet the demand that above-mentioned virtual image expression realizes.The present invention can utilize the image resource of lesser amt, and the independent assortment virtual image producing a myriad of selects to represent the virtual image of individual for user, and expresses individual's emotion or emotion with virtual image facial expression.The present invention is applicable to different types of two and three dimensions virtual image, including face, whole body, animal, casper and abstract emoticon face.

Description

Virtual image for independent assortment creation generates and uses the method and system of expression

Technical field

The present invention relates to area of computer graphics, particularly to a kind of for independent assortment creation virtual image generate and Use the method and system of expression.

Background technology

Darwin mentions in " expression of human and animal " book of his publication in 1872, and the mankind have following spy Different expression: painful and sobbing, demoralized, worried, grieved, dejected, disappointed, happy, mental exertion, love Feelings, softheartedly, worship, recall, meditate, severe emotion, sulky, be determined, hate with indignation, disdain, scorn, Detest, guiltiness, pride, isolated, restrain oneself, affirmation and negation, surprised, startled, frightened, frightened, Oneself note, ashamed, modest, blush (BJ University Press, Zhou Bangli translates).Darwin believe the mankind (with Animal) evolve express emotion and identify expression ability, be make the mankind (animal) can generate social organization from And strengthen the important means of species viability power；Expression and spoken and written languages act on the existence of human society, development in the lump Basis.The mechanism that psychologist, physiologist generally believe facial expression is relevant to the change of the muscle of different parts, But the most still cannot resolve wherein secret set up complete model completely, because facial expression muscular tissue System at least following 18 kinds of muscle (point left and right having): auricularis anterior, buccinator muscle, depressor supercilii, depressor anguli oris, fall Lip flesh, depressor septi, frontalis, levator anguli oris, levator labii superioris, mentalis, nasalis, orbicularis oculi, mouth wheel circle Flesh, platysma, procerus, Albinus' muscle, zygomaticus major, zygomaticus minor；A combination thereof effect is extremely complex delicate.Based on This, as the field that a comparative maturity and commercialization are strong, computer generates doing of image circle (CGI) Method is, by motion capture, the expression of picture catching true man scapegoat, then expression is reappeared fixing animation people Personage's face expression is generated on thing moulding face rather than through histrionic muscles meat tissue system model algorithm.CGI Boundary is not required to as quantity-unlimiting, and the face generation expression that non-predetermined is fixing, is such as numerous in the random masses of true man Individual face generates facial expression image.At the beginning of the Internet is born, exist for " feelings using text as the epoch of online exchange Sense symbol " (emoticon, the facial expression that word punctuation mark is drawn, such as:-) also known as " smile's symbol " smiley). Today, emoticon was more graphical and personalized, but its multiformity and emulation property the most far from be enough to distinguish individual user, User is allowed to bring the personal virtual image as oneself.

Computer virtual image is increasingly being used in online social environment, becomes formation user and brings sense into intimate One important component part of sense.Computer virtual image technology can be led to as the field of a market relative maturity Cross the different virtual image parts organ of independent assortment to create the different virtual image faces of a myriad of, but this A little virtual image faces lack the function doing different facial expressions.The most well-known User Defined virtual image system Such as simulate life (Sims), World of Warcraft (World of Warcraft), Tencent QQ etc., their free group Close virtual image and all can not do different facial expressions.In addition, traditional police faces combination sketch atlas system What system was generated is also the expressionless face of stationary plane.These precedent neither ones can be the virtual image of independent assortment Face generates multiple facial expression, although multiple facial expression can strengthen the Consumer's Experience of product and the social activity of user Experience.

Summary of the invention

It is an object of the invention to prior art and cannot generate multiple face flexibly for the virtual image face of independent assortment The defect of expression, thus a kind of side that can generate multiple facial expression for the virtual image face of independent assortment is provided Method and system.

To achieve these goals, the invention provides one and be implemented at least one calculating device, described calculating fills Put including at least a processor, memorizer and communication platform interface, generate the method specifying expression for virtual image, Described virtual image includes but is not limited to two dimension or the personage of three-dimensional, animal, casper or abstract emoticon Face, face's organ or local face acquired in the method come from one and come for combination producing difference virtual image Representing the set of multiple user, the method comprises the following steps:

Step 1, the reception first information, the described first information is relevant to the virtual image representing user；

Step 2, based on the first information from above-mentioned set obtain one or more faces organ or local face；

Step 3, based on above-mentioned face organ or local face, generate represent user virtual image basis synthesis face；

Step 4, receive the second information, the appointment expression that described second information to be shown about virtual image；

Step 5, based on the second information from above-mentioned set obtain one or more faces organ or local face；

Step 6, based on basis synthesis face and step 5 obtained by face's organ or local face, for virtual image Generate the synthesis face showing above-mentioned appointment expression.

In technique scheme, the described first information include following at least one, from multiple default basis synthesis faces The selection information of the basis synthesis face needed for middle appointment, or one or more attribute letter relevant with virtual image Breath.

One or more relevant with virtual image attribute letter in technique scheme, included by the described first information Breath, including following at least one, sex, age, race, occupation.

In technique scheme, face's organ or the local face of described acquisition have related metadata.

In technique scheme, described with face's organ or the metadata that associates of local face include following at least one: Expression information, identity information, Viewing-angle information, profile information, positional information.

In technique scheme, face's organ of described acquisition or local face, the metadata being associated based on it.

In technique scheme, described second information include following at least one:

User inputs, it is intended that the expression that its virtual image to be shown；

The parameter that one or more applied environments residing for virtual image are relevant；

The information that can act on virtual image sent from social networks.

In technique scheme, also include that storage can show and specify the independent assortment virtual image of expression or band expression to close Becoming face, this storage information is applied for other.

In technique scheme, described virtual image band Expression synthesis face is rendered and is stored as picture, described figure Sheet is static images or the dynamic picture with expression animation.

In technique scheme, it is stored as reappearing this virtual image by described virtual image band Expression synthesis face Data message, described data message include face's organ used by virtual image or local face identification information.

In technique scheme, described virtual image also includes limbs organ, the limb action symbol of described virtual image Close described appointment to express one's feelings.

In technique scheme, described collection is combined into virtual image organ data storehouse.

In technique scheme, also include the step setting up or expanding virtual image organ data storehouse.

In technique scheme, user includes in the information sent to other users doing the virtual image expressed one's feelings.

In technique scheme, user makes for other users and can do the virtual image expressed one's feelings, and by this virtual shape As being sent on social networks as sharing.

In technique scheme, the second information received in step 4 is an expression sequence information, in step 6 In generate for virtual image and to show the synthesis face of expression in this expression sequence.

Present invention also offers one to be implemented at least one and calculate device, described calculating device is including at least at one Reason device and memorizer, the method setting up or expanding the virtual image organ data storehouse that band is expressed one's feelings, this data base is for combination Generating different virtual image and represent multiple user, the method step includes:

Step 1, the virtual image obtaining band espressiove or the original image set of partial virtual image, for described former Beginning image identification its expression and affiliated virtual image, the form that described original image uses includes but not limited to one or many Graph description format including bitmap format, vectogram form, or threedimensional model form；

Step 2, generate face's organ or the local image of face and metadata, metadata bag based on described original image Include expression information.

In technique scheme, the metadata of described face organ or local face has identity information, described identity Information is based on virtual image belonging to its ingredient original image.

In technique scheme, described face organ or local face metadata include following at least one:

Two dimensional image Viewing-angle information, or give the metadata described two dimension face's organ without Viewing-angle information or local One, face acquiescence Viewing-angle information；

3-D view profile information, or give the metadata described three-dimensional face's organ without profile information or local One, face acquiescence profile information.

In technique scheme, described face organ or local face metadata also include positional information, described position Information is for this face's organ or face location in whole virtual image, local.

In technique scheme, also include marking at least one attribute information for described face organ or local face, Described attribute information include but not limited to identification information, classification information, colouring information, age information, gender information, Ethnic information, occupational information.

Invention further provides and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, the method controlling the expression of user's virtual image, described virtual image is the virtual image that independent assortment generates:

Based on one or more to virtual image residing for the relevant parameter of applied environment, the expression of trigger condition will be met It is set as the expression of user's virtual image.

Present invention also offers and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, at least one of which device also comprises localizer or touch screen, the method controlling the expression of user's virtual image, Described virtual image is the virtual image that independent assortment generates, including:

Step 1, reception specify expression sequence, and described expression sequence is an expression, or user is by slip localizer Or streak multiple expressions that touch screen selects；

Step 2, the expression of user's virtual image that the expression meeting trigger condition in above-mentioned expression sequence is set as.

Invention further provides and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, the method searching user with virtual image, described virtual image is the virtual image that independent assortment generates, Including:

Avatar information is specified in step 1, reception；

Step 2, return at least one user profile, its virtual image parts metadata with specify virtual image parts unit Data meet matching condition.

Present invention also offers one to be implemented at least one and calculate device, described calculating device is including at least at one Reason device, memorizer and communication platform interface, generate the system specifying expression, described avatar pack for virtual image Include but be not limited to two dimension or the personage of three-dimensional, animal, casper or abstract emoticon face, this system institute The face's organ or the local face that obtain come from one for combination producing difference virtual image to represent multiple user's Set, this system includes:

First information receiver module, is used for receiving the first information, the described first information and the virtual image representing user Relevant；

First data acquisition module, obtains one or more faces organ or local based on the first information from above-mentioned set Face；

Basis synthesis face generation module, based on above-mentioned face organ or local face, generates and represents the virtual of user Image basis synthesis face；

Second information receiving module, for receiving the second information, described second information is intended to show about virtual image Appointment expression；

Second data acquisition module, obtains one or more faces organ or local based on the second information from above-mentioned set Face；

Band Expression synthesis face generation module, obtained by basis synthesis face and described second data acquisition module Face's organ or local face, for virtual image generate show above-mentioned appointment expression synthesis face.

In technique scheme, the metadata that face's organ of described acquisition or local face are associated based on it.

The information that can act on virtual image sent from social networks.

In technique scheme, also include a memory module, can show that the independent assortment specifying expression is empty for storing Intending image or band Expression synthesis face, other application need not described different for combination producing when using this storage information Virtual image represents the set of multiple user.

In technique scheme, described virtual image band Expression synthesis face is rendered and is stored as by described memory module Picture, described picture is static images or the dynamic picture with expression animation.

In technique scheme, described virtual image band Expression synthesis face is stored as weighing by described memory module The now data message of this virtual image, described data message includes face's organ used by virtual image or local face Identification information.

In technique scheme, also include the module setting up or expanding virtual image organ data storehouse.

In technique scheme, the second information that the second information receiving module is received is an expression sequence information, band Expression synthesis face generation module is that virtual image generates the synthesis face of expression in this expression sequence of display.

Invention further provides one to be implemented at least one and calculate device, described calculating device is including at least at one Reason device and memorizer, set up or expand the system in the virtual image organ data storehouse that band is expressed one's feelings, and this data base is for combination Generating different virtual image and represent multiple user, this system includes:

Original image acquisition module, for obtaining virtual image or the original graph of partial virtual image of band espressiove Image set closes, and identifies its expression and affiliated virtual image, the form bag that described original image uses for described original image Include but be not limited to one or include more bitmap format, vectogram form in interior graph description format, or threedimensional model lattice Formula；

Image and metadata generation module, based on described original image generate face's organ or local face image and Metadata, metadata includes expression information.

The present invention has reoffered and a kind of has been implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, controls the system of the expression of user's virtual image, and described virtual image is the virtual image that independent assortment generates, This system includes:

Expression setting module, based on one or more with virtual image residing for the relevant parameter of applied environment, general is satisfied The expression of trigger condition is set as the expression of user's virtual image.

Present invention also offers and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, at least one of which device also comprises localizer or touch screen, controls the system of the expression of user's virtual image, Described virtual image is the virtual image that independent assortment generates, and this system includes:

Expression receiver module, is used for receiving appointment expression sequence, and described expression sequence is an expression, or user is led to Cross slip localizer or streak multiple expressions that touch screen selects；

Expression setting module, for being set as the virtual shape of user by the expression meeting trigger condition in above-mentioned expression sequence The expression of elephant.

Present invention also offers and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, searches the system of user with virtual image, and described virtual image is the virtual image that independent assortment generates, This system includes:

Virtual image receiver module, receives and specifies avatar information；

Virtual image searches module, returns at least one user profile, its virtual image parts metadata and appointment void Intend image parts metadata and meet matching condition.

It is an advantage of the current invention that:

The method and system of the present invention need not motion capture, the face of picture catching individual user, it is possible to utilizes relatively The image resource of small number produces the independent assortment virtual image of a myriad of, selects to represent the virtual of individual for user Image and express individual's emotion or emotion with virtual image facial expression.

Accompanying drawing explanation

Fig. 1 is the schematic diagram that in face, loci divides；

Fig. 2 is the schematic diagram of image information and metadata；

Fig. 3 is that the virtual image of the band espressiove of the present invention generates the virtual image that method generation display appointment is expressed one's feelings The schematic diagram of basic process；

Fig. 4 is the flow chart of the virtual image generation method of the band espressiove of the present invention；

Fig. 5 is the schematic diagram that the face's organ employed in an embodiment selects interface；

Fig. 6 is the example schematic that virtual image band expression organic image is replaced；

Fig. 7 is the schematic diagram of three dimensional structure's model；

Fig. 8 is the schematic diagram of the threedimensional model generating band expression；

Fig. 9 is the flow chart of method setting up or expanding virtual image organ data storehouse of the present invention；

Figure 10 is the schematic diagram of the virtual image expression instrument using the method for the present invention to be realized in an embodiment；

Figure 11 is in one embodiment, and generating for independent assortment virtual image of the present invention specifies the system of expression to make For social data server or the schematic diagram of the back-end system of game server；

Figure 12 is in one embodiment, and generating for independent assortment virtual image of the present invention specifies the system of expression to make Schematic diagram for independent ISP；

Figure 13 is in one embodiment, the present invention for independent assortment virtual image generate specify expression system with Virtual image organ creates the system schematic diagram respectively as independent ISP；

Figure 14 is the schematic diagram of the virtual image expression personal letter using the method for the present invention to be realized in an embodiment.

Detailed description of the invention

In conjunction with accompanying drawing, the invention will be further described.

Concept definition

In order to make it easy to understand, first some concepts involved in the present invention are done unified explanation.

Client: client (client-side device) can be such as personal computer PC (Personal Computer), notebook computer, panel computer, mobile phone, smart mobile phone are at interior terminal unit；Can also be eventually Client modules in end equipment, such as: web browser (web browser) client, instant messaging (instant Messenger) applications client etc..

Loci: loci (alignment origin) refer to for multiple images are overlaped and guarantee each figure it Between the consistent common origin of relative bearing；Printing circle be similar to concept be alignment mark (registration mark) especially For four-color process.

Virtual image: virtual image (avatar) is for representing personal user in the Internet, internet, applications, And show other users figure image, can only containing head portrait or hold concurrently containing health, limbs, hair style, dressing, join Decorations.Virtual image also includes casper virtual image, animal virtual image and the most abstract approximation emoticon Virtual image；In embodiments herein, illustrate as a example by mankind's virtual image.

Face's organ: face's organ (facial feature) refers to a certain organ in virtual image face, with face As a example by, the face's organ on face includes shape of face, eyebrow, eyes, nose, face, ear etc..

Locally face: locally face (partial face) refers to a part for the complete face of virtual image, and it is some The combination of individual face organ, as by the local face of face, nose and the skin composition around both.

Face parts: face's organ and local face are referred to as face parts (facial components).

Limbs organ: limbs organ (body parts) refers to the virtual image a certain organ in addition to face, with people As a example by, trunk, arm, hands, lower limb, foot etc. belong to limbs organ.

Independent assortment virtual image: independent assortment virtual image (free-combination avatar) refers to select respectively Face's organ, locally face or limbs organ (selecting eyebrow, eyes, nose, face, shape of face the most respectively), Then the virtual image combined.

Expression: expression (expression) refers to for showing emotion (emotion), the face face of emotion (mood) Portion's state, coordinates limb action or implication tone sometimes.

Virtual image organ data storehouse: virtual image organ data storehouse (avatar features database) refers to a use In component set and the relevant information of the different virtual image of combination producing, its content components can be face's organ, Locally face or the image information of limbs organ, its relevant information can be the metadata of all parts, attribute Information etc..

It is above the explanation to related notion involved in the present invention.The most all as a example by the mankind, The virtual image how method of the present invention generates band espressiove is described.The method using the present invention equally may be used To generate the virtual image of other kinds of band espressiove, such as casper or animal virtual image.

Face's organ or local face acquired in the method for the present invention come from one for the different virtual shape of combination producing As representing the set of multiple user.In a preferred embodiment, this set is a virtual image organ data Storehouse；First being described in detail this data base, how the method then putting off until some time later the bright present invention uses this data base.

In one embodiment, virtual image organ data storehouse contain described face organ, described local face with The information of described limbs organ.These deposits of faith include that two classes, a class are image informations, and another kind of is metadata； Corresponding relation is there is between image information with metadata.Fig. 2 gives the schematic diagram of image information and metadata.

Described image information describes face's organ or local face or the shape of limbs organ.Image information can be Two dimensional image form, such as bitmap format, or vectogram form, or other available graph description format, it is also possible to It it is threedimensional model form.

Described metadata is the relevant information for describing correspondence image information.Described metadata includes expression information, Expression information can reflect the expression of corresponding image, and the expression as corresponding to a face opened is " laugh ", Expression corresponding to one face being slightly upturned is " smile ".Corresponding to some face's organ or local face Expression can have multiple, if a face closed can be " sad " or " nervous ".Expression information can also be Sky, this represents the face's organ representated by corresponding image or local face or limbs organ is not expressed one's feelings with specific It is associated, as " ear " generally would not be associated with a certain specific expression.

Described metadata may also include identity information, and described identity information is for identifying the face representated by correspondence image Which virtual image organ or local face or limbs organ belong to.Identity information is identical or the face that matches Organ or locally face or limbs organ, more can alternatively use and not change the identity of virtual image.If one Individual local face comprises the organ combination from different virtual images, and its identity information comprises the identity of its indivedual organs Information, i.e. identifies its virtual image belonging to indivedual organs.

Described metadata may also comprise the Viewing-angle information of two dimensional image, and described Viewing-angle information is for describing corresponding two The observation angle of dimension image.Obviously, same organ is looked up, side is cut with facing observed two dimensional image The most different.Identical or approximation visual angle image more can be combined.If there is an acquiescence visual angle, meet The image at this acquiescence visual angle, its metadata can be without Viewing-angle information.

Described metadata may also comprise the profile information of 3-dimensional image model, and 3D shape of face model is according to organ nominal contour Hollowing out, each 3D organ model meets the type organ nominal contour, it is ensured that can be implanted in 3D shape of face model, And (seeing Fig. 7) can be exchanged with same type 3D organ model.If there is an acquiescence organ nominal contour Combination, meets the model of this default standard profile, and its metadata can be without profile information.

Described metadata may also include positional information, and described positional information is for describing corresponding image in whole void Intend location in image, it is also possible to comprise zoom scale and the anglec of rotation etc..As a kind of preferred implementation, Described positional information comprises the type of loci and the type loci in image, or outside image boundary, Position (generally, this loci is with the distance in the image upper left corner).There are multiple images of same type loci, press After above-mentioned zoom scale and above-mentioned anglec of rotation normalizing, when overlaping and make the type loci align, this The relative bearing relation of a little images is correct, it is possible to show the most in the lump.

Physiology mechanism according to human body, applicants have observed that: after height and head size normalizing, have on human body Several key positions, such as two centers, the umbilicuss of human body of face, for different appearance, the people of different building shape, Relative bearing relationship change between each organ and these key positions is little；A kind of preferred implementation is taken These key positions are as loci.See Fig. 1, the location of face's organ, based on face mask and a cross, This cross, by being formed by nose-mouth with by two straight lines of left eye-right eye, is substantially stationary；With above-mentioned ten The cross point (i.e. two centers) of word is as loci, and face's organ of the people of different appearance is for this loci Relative bearing is also substantially stationary.So for different shapes of face, face's organ, using this loci as scheming As the loci of aligned display contributes to combining different face's organ or local face.Same, using umbilicus as people The loci of body, contributes to combining different limbs organ.Described face organ or local face (are referred to as face portion Part) image, or limbs organic image, relevant location information need to comprise and have with these face parts or limbs organ The type of the loci of relative bearing relation, and the position that this loci is in image or outside image boundary.Right In other virtual images personalized, such as animal, it would however also be possible to employ similar method unifies its loci, thus will Overlapping displaying after the alignment of multiple organic images.

Described metadata may also include attribute information, and attribute information includes but not limited to the one in following information or many Kind: identification information, classification information, colouring information, age information, gender information, ethnic information, occupational information. Identification information therein is for identifying corresponding image, and described mark has uniqueness, as represented " nose " by one Image identification be " 100001 ".Classification information is for describing the face parts representated by correspondence image or limbs organ The type of (hereafter referred to collectively as organ parts), is " nose " as marked the type of a certain face organ.Colouring information, For describing the color of the organ parts representated by correspondence image, such as " brown ", " black ", " blue " (eyes) Deng.Age information is for describing the age of the virtual image of the organ parts institute subordinate representated by correspondence image, with body The virtual image of existing different age group is in the apparent difference of organ parts.Gender information is used for describing correspondence image institute The sex of the virtual image of the organ parts institute subordinate represented, to embody the virtual image of different sexes organ parts Apparent difference；As man, female have more significantly difference on the face such as eyes, face organ；Certainly certain The gender information of a little organ parts allows also to be neutral, and this means that this organ men and women is all suitable for.Ethnic information is used In the race of the virtual image of the organ parts institute subordinate described representated by correspondence image, to embody the most agnate void Intend image in the apparent difference of organ parts.Occupational information is for describing the organ parts institute representated by correspondence image The occupation of the virtual image of subordinate, to embody the virtual image of different occupation in the apparent difference of organ parts, as Athlete is the thinnest.

Described virtual image organ data storehouse includes the information of polytype organ parts, type refer to " eyes ", " nose ", " face " etc.；Described virtual image organ data storehouse includes same type of organ parts many Individual image and the metadata of association.As a example by " face ", may comprise in described virtual image organ data storehouse There is the information of hundreds of kind different " face ", these " faces " or have different identity (as outward appearance is presented as Thick lip or thin lip), or have different colors (having smeared different colours lip pomade as outward appearance is presented as), or have difference Expression (as smile or laugh).When identity information is identical or matches, when two dimensional image Viewing-angle information or three-dimensional Image outline information is identical or matches, when expression information meets the expression required for application, and these images are permissible The virtual image of the band espressiove required for applying for combination producing.When attribute information meets looking into by application settings Asking or screening conditions, these images can be that application provides the candidate that can be combined as virtual image.

Various information in described virtual image organ data storehouse are when storage, as a kind of preferred implementation, and can Shape information (it externally shows as graphic file) and out of Memory are stored separately in different data acquisition systems. The relation between data in multiple data acquisition systems can be associated by modes such as identification information, index, pointers. Advantage of this is that and be prone to do data query, index, update, transmit and store.

It is above the description in virtual image organ data storehouse；How it sets up or the method that expands will additionally be done below Explanation.First, the virtual image using virtual image organ data storehouse to generate with expression is illustrated how.

In one embodiment, virtual image organ data storehouse is comprised face's organ or local face (are referred to as Face parts) information at least include the shape of these face parts, position, classification, identity, visual angle, expression etc. Information.Citing, virtual image organ data storehouse comprises 6 male's identity of same view angle and 6 women 5 kinds of face's organs of basic face (first not considering expression) of identity, eyebrow, eyes, nose, mouth and shape of face are (i.e. Each type organ has 6 male and 6 women), the most different；The most only eyebrow, eyes, mouth for It is default that different expressions has different shapes, eyebrow, eyes, the organ of mouth these three type to be respectively arranged with identical 18 kind Expression.The organ of masculinity and femininity can not exchange, but upper between male face organ, between female face organ State same type organ can exchange, because its visual angle is unified.Do not consider that other virtual image appearance attribute is such as sent out Type, the colour of skin etc., this data base can combination producing 2 × 6 × 6 × 6 × 6 × 6=15552 different men, female preset Basis synthesis face, each corresponding different combination virtual image identity, 5 kinds of organs of this combination virtual image (eyebrow, eyes, nose, mouth, shape of face) is respectively from the above-mentioned same sex other basis face organ and (does not first examine Consider expression)；The combination of this organ is equal to the identity information of this combination virtual image.Further, the virtual shape of each combination As identity, count its basis synthesis face in, be respectively arranged with 18 kinds of expressions (because each band expression organ has identical 18 Plant and preset expression).Above-mentioned virtual image organ data storehouse, therefrom selects for social network user to represent the virtual of oneself Image and enable its virtual image to show these 18 kinds expressions.

Based on this virtual image organ data storehouse, the virtual image implementation method of the band espressiove of the present invention generates display Specifying the virtual image of expression, with reference to Fig. 3, its ultimate principle is: generate a basis synthesis face for user, so Choose from virtual image organ data storehouse according to the expression to be shown afterwards and meet the face parts of some condition and replace Corresponding face parts in generation basis synthesis face, generate display and specify the virtual image of expression.(herein, " substitute " Refer to display band specify expression face parts and do not show the basic face parts of same type, it is achieved mode it may be that But being not necessarily, basis face parts data " being replaced " on the synthesis face generated is that expression face specified by band The data of parts.) generally, generate and represent the step of user's virtual image basis synthesis face at this user base of display Complete during face (such as " poker-faced "), the when that this can being to allow user select the virtual image representing oneself, Can also be, when any user virtual image should " poker-faced " when；Generating display specifies expression to close The step becoming face completes when allowing the display of any user's virtual image specify expression.Different tables before and after having The virtual image that feelings change " does and expresses one's feelings " visual effect, is only the distinctest effective.

Fig. 6 is the example schematic that virtual image band expression organic image is replaced, and its data are above-mentioned virtual shape Part data as organ data storehouse.Wherein A, B, C, D, E, F are the body of 6 entirely different virtual images Part, they be respectively arranged with common, proud, angry, sad, ridicule, close, helpless, excited, disappointed, hesitate, put The expression that the heart etc. are different；As a example by virtual image A, when showing multiple different expression, the eyebrow of A, eye, mouth are equal Do organic image replacement.And two virtual images of G and H are by these 6 virtual image face organs of aforementioned A-F Combine.Combination virtual image G metadata identity information be: eyebrow from A, eye from D, nose from E, Mouth is from B, and shape of face is from A, and hair is from F；The metadata identity information of virtual image H is: eyebrow from C, Eye is from C, and nose is from F, and mouth is from E, and shape of face is from C, and hair is from A.During virtual image G to be shown, In step S101-step S103, from virtual image organ data storehouse, obtain A-eyebrow-common, D-eye-common, E- Nose (amimia), B-mouth-common, A-shape of face (amimia), F-hair (amimia), thus generate basis synthesis Face G-is common；In step S104, citing specifies expression for " proud "；In step S105, acquisition A-eyebrow-obtain Meaning, D-eye-proud, B-mouth-proud；In step S106, replace the A-eyebrow in basis synthesis face-common, D-eye-common, B-mouth-common, ultimately generate and show that the synthesis face G-of proud expression is proud.

The handling principle of three-dimensional avatars is essentially identical with two-dimensional virtual image, but is implanted to shape of face model to replace In organ model realize.In one embodiment, virtual image is threedimensional model.Such as Fig. 7, contain one 3D shape of face model (containing ear) after organ, the face face different with two groups have been hollowed out according to organ nominal contour 3D organ model set (left and right eyebrow, left and right eyes, nose, face)；It is implanted in shape of face model, Above-mentioned two groups of organ model will form two different virtual images, image 1 and image 2 respectively.One virtual image 3D organ model set be to comprise multiple expression (if " laughing at " and " close " is two groups of 3D organ model), The different expression threedimensional models of this virtual image are formed in replacement 3D organ model mode.Such as Fig. 8, contain above-mentioned " laughing at " and " close " of two virtual images two kinds expression threedimensional model (sectional drawing is front, side, oblique 45 degree of angles). Because the cavity in shape of face model all hollows out according to organ nominal contour, the 3D organ model with type organ can be mutual Implant with changing, catch up with state that two-dimensional virtual image compound mode is the same can the three-dimensional shape of combination producing different identity As (face such as taking the face of aforesaid image 2 to substitute image 1 forms new composite figure 3).Implement at this In example, a three-dimensional avatars organ data storehouse, therefrom select for social network user to represent the three-dimensional empty of oneself Intend image and enable its three-dimensional avatars to show multiple expression.

Now, as a example by two-dimensional virtual image, describe in detail and generate, into virtual image, the method specifying expression.Reference Fig. 4, the method specifically includes following steps:

Step S101, the reception first information, the described first information is relevant to the virtual image representing user；

In step S101, the described first information include following at least one, from multiple default basis synthesis faces The selection information of the basis synthesis face needed for appointment, or the one or many attribute information relevant with virtual image.Institute State the attribute information that one or many included by the first information is relevant with virtual image, including following at least one: sex, Age, race, occupation.

In one embodiment, although user cannot travel through all default bases that he or she can select realistically and close Becoming face, user can select face's organ or local face (to be referred to as face by such as the interface of Fig. 5 one by one Parts).The type (such as eyebrow, eyes, nose, face, shape of face) of face parts is selected by position button, Specific face parts (the choosing eye in such as Fig. 5 is selected the most again from the list showing the different face parts of the type Eyeball, other type organ is constant).The face organ type that user is unselected, it is possible to use acquiescence organ set in advance. The selection information aggregate of above-mentioned all types of face parts, based on synthesize face select information, the most above-mentioned first letter Breath.

In another embodiment, according to instruction, the physiological structure in conjunction with face selects synthesis face in basis to be generated Hole.Wherein, described instruction is used for describing virtual image attribute selection condition to be selected, as about sex, year Age, race, the screening conditions of occupation, it is also possible to be other screening conditions, such as eyes sizes values.Described instruction Can be inputted by user, it is possible to use a default instruction set in advance.If described instruction simply show whole face The screening conditions of a part of face parts in hole, the most also can mend according to the physiological structure of face automatically The screening conditions of neat remainder, these screening conditions can be attribute default set in advance.Above-mentioned attribute sieves Select conditional order, as attribute information relevant with virtual image, the most above-mentioned first information.

In another embodiment, above-mentioned for specifying the selection information of basis synthesis face and above-mentioned for screening The attribute information of virtual image, and be stored among the above-mentioned first information, such as with screening mode supplement selection information it Not enough.

In a preferred embodiment (" make for good friend and share band expression virtual image "), user can be other User makes and can do the virtual image expressed one's feelings, and is sent on social networks this virtual image as sharing；The One information is, this virtual image, represents the user being produced virtual image.Make for good friend and share band expression void Intend image, as a complete embodiment, will the most individually illustrate.

Step S102, based on the first information from a face's organ being available for combination producing difference virtual image or local The set of face obtains one or more faces organ or local face.

Described face organ or local face have related metadata.Described with face's organ or local face associates Metadata include following at least one: expression information, identity information, Viewing-angle information, positional information.Described obtain The metadata that the face's organ taken or local face are associated based on it.

In one embodiment, combination producing difference virtual image it is available for described in represent face's organ of multiple user Or the set of local face, its a kind of implementation is virtual image organ data storehouse.Above-mentioned steps S101 embodiment Selected in the component options of all types of face parts may come from this organ data storehouse tape identification information, this mark Knowledge information is as a part for above-mentioned selection information so that this step can obtain corresponding face based on this identification information Hole parts.Virtual image organ data storehouse includes image and the metadata of described face parts, and this metadata includes table Feelings information, identity information, Viewing-angle information, positional information etc..Above-mentioned steps S101 embodiment has with virtual image The attribute information (i.e. screening conditions) closed, can mate with the attribute information metadata in organ data storehouse, obtains Meet the metadata of screening conditions and the virtual image face parts of correspondence.

In another embodiment, step S102 also includes acquired all faces organ or the letter of local face Breath does reasonability and verifies with legitimacy, such as can have nose, it is impossible to nude etc..

Step S103, based on above-mentioned face organ or local face, generate represent user virtual image basis synthesis Face；

Wherein, when the shape information of face's organ or local face is combined, can be according to face's organ or office The positional information of portion's face.

In one embodiment, with the loci included in positional information, zoom scale and anglec of rotation phase As a example by the information of pass, when the image of multiple face parts is combined, first zoom scale and the anglec of rotation are returned One changes, the most one by one by the image that not yet positions according to the position of its loci this loci and oriented image Same type loci alignment (loci can have different types of), until all of image all complete location.

Also need to consider that the figure layer between face parts to be combined blocks pass when by the image sets of face parts altogether System.As a example by eyes and shape of face, when combination, eyes figure layer (should assume draw above on shape of face figure layer Layer blocks following figure layer), then according to correct figure layer blocks order display synthesis face.

Step S104, receive the second information, the appointment expression that described second information to be shown about virtual image；

It is one or more that described second information includes but not limited in following information: user inputs, it is intended that it is virtual The expression that image to be shown；The parameter that one or more applied environments residing for virtual image are relevant；From social activity The information that can act on virtual image that network sends.

Above-mentioned three kind of second information not mutual exclusion, concrete second information may both input from user, also through Social networks sends, and is also the parameter that reflection applied environment is relevant.

In one embodiment, user input specified expression (above-mentioned second information) can be with above-mentioned virtual shape As the expression information metadata in organ data storehouse matches, thus filter out the face parts showing this expression.On Stating user's input can be to select to describe the word tag of a certain concrete expression by drop-down menu, as " happily ", " sad " etc., it is also possible to be by clicking on the pattern mark selecting a certain concrete expression, asDeng, Can also be by input through keyboard emoticon or other character string such as ":-) ", " :-(" etc., or other user is mutual Mode.

(seeing " the virtual image expression instrument " in Figure 10) in a preferred embodiment, user is pointed to by slip Device or streak touch screen and select a sequence expression mark, wherein meeting the expression of trigger condition becomes described second letter Breath, the expression that this user's virtual image to be shown.Above-mentioned trigger condition is it may be that stop localizer or touch screen During operation, (such as unclamp mouse mouse-up, finger leaves touch screen) localizer or finger are positioned at certain expression mark； Can also be that localizer or finger rest on certain expression mark long enough time；Can also be that other user is mutual, Such as double click or touch control gesture etc..In another embodiment, above-mentioned expression sequence can trigger multiple to be shown Expression, arrange display in order, such as " laugh at-cry-laugh at-cry-laugh at ", its implication is " not knowing whether to laugh or cry " (between tears and laughter)。

(see " the virtual image expression personal letter " in Figure 14) in a preferred embodiment, send out personal letter user with Drop-down menu or alternate manner select expression, as the second information, the user receiving personal letter receive this second information. In one embodiment, the user sending out personal letter and receipts personal letter is social network user, and above-mentioned second information is from society Hand over network.In one embodiment, above-mentioned second information is the parameter that applied environment is relevant, and its implication is, has band The personal letter of this virtual image expression is available for reading.Virtual image expression personal letter, as a complete embodiment, will The most individually illustrate.

In another embodiment, in the application showing virtual image, ambient parameter information also can certain time The expression that reflection virtual image to be shown, such as in a social networks game, according to the development of game, user The virtual image of oneself or good friend can demonstrate that " excited ", " dejected " etc. are expressed one's feelings；Above-mentioned second information is table Feelings parameter, acts on, and represents the virtual image of some user.

In a further advantageous embodiment (" strategy of automatically expressing one's feelings "), further, based on above-mentioned ambient parameter information, The expression meeting trigger condition is set as the virtual image expression of user oneself or other users.Preferred as one Implementation, provides the user the expression strategy of a set of acquiescence and such as expresses one's feelings trigger condition list, and user can from The interface of definition expression strategy.Automatically express one's feelings strategy, as a complete embodiment, will the most individually illustrate.

In another embodiment, the information from social networks also can reflect the virtual image of user oneself or good friend The expression to be shown, such as, when good friend logs in this social activity networking, and the virtual image of this good friend and user oneself is equal Can show and can reflect the expression of relation between two people, such as " smile ", " smile add wave ", " disdain to add and turn around " etc.； Above-mentioned second information is the virtual image expression information from social networks.

Step S105, based on the second information from above-mentioned set obtain one or more faces organ or local face；

In one embodiment, the acquisition mode of described face organ or local face is similar to side described in step S102 Formula, which obtains the face parts of basis synthesis face, and the second information is i.e. expressed one's feelings by this step, as face Parts expression information metadata screening conditions.In a preferred embodiment, based on the face acquired in step S102 Parts, obtain in order to the face parts substituted, and metadata identity information is identical with Viewing-angle information between the two or phase Coupling, meets the above-mentioned expression to be shown in order to the face parts metadata expression information substituted.

In one embodiment, when the face component set meeting above-mentioned screening conditions also needs to further screen, Other meta-data attribute information screening conditions such as classification information, colouring information, age information, Zhong Zuxin can be added Breath, occupational information etc..

Step S106, based on basis synthesis face and above-mentioned face organ or local face, generate aobvious for virtual image Show the synthesis face that above-mentioned appointment is expressed one's feelings.

In one embodiment, based on the face parts obtained before regenerate one new, expression is specified in display Synthesis face, its mode and step S103 are basically identical rather than on the basis of the synthesis face the most generated " replace " some face parts data.

In another embodiment, it is also possible to by replacing face parts, generate the conjunction showing above-mentioned appointment expression Become face.When realizing face part replacement, need to remove the face parts that will be substituted face from having synthesized, Being inserted into the face parts substituted, positional information and the shape information of described face parts is used in this operation, May need based on the zoom scale in positional information and the anglec of rotation, face image of component to be first normalized.This The operation of face part replacement also may change each face parts hiding relation in synthesis face, is therefore replacing After may also need to adjust the figure layer relation of each face parts in synthesis face, it is ensured that correctly show.

In another embodiment, the image of face parts and relevant information can also change this face by calculating The attributes such as the shape of parts, color, the colour of skin, make required for the display application of synthesis face or the user of virtual image Effect；One example is " opening eyes wide ", and the width of eye image is constant but is highly multiplied by a multiplying power (such as 1.2 Times)；Another example is " painting lip pomade ", changes the color of lip in image；Another example is to pass through image Affine transformation (affine transformation) change its Visual Angle in Perspective.

More than complete to generate the detailed description specifying expression method into virtual image.Several the method be will now be described Derivative application, relevant to virtual image expression with several, but do not rely on the application of above-mentioned steps S101-S106.

The synthesis face that above-mentioned steps S106 generates, can be stored as display and specify the virtual image synthesis face of expression, Do not have complete virtual image organ data storehouse still can show afterwards.A kind of mode is by described virtual image band table Feelings synthesis face renders and is stored as picture, and described picture is static images or the dynamic picture with expression animation. Described virtual image band Expression synthesis face is stored as reappearing this virtual image expression by another mode Data message, described data message can include the identification information set of each face parts used by virtual image, such as word Symbol string " 1001,2001,3001,4001,5001,6001；Expression: 1 "；Afterwards, resolve this number evidence and obtain After taking its face parts picture pointed to and relevant location information, can be by the method described by step S103 Above-mentioned picture synthesizes, and described data message is reappeared the virtual image for band espressiove.

The expression of people is in addition to can showing by face, in some cases, it is possible to shown by limb action Come.Therefore, in another embodiment realized on the basis of aforesaid plurality of embodiment, the present invention with table The virtual image implementation method of feelings also includes the virtual image limb action generating band espressiove.The method is based on Virtual image organ data storehouse also includes the information of limbs organ, the information of described limbs organ include shape information, Positional information, classification information, expression information, identity information；May also include colouring information, age information, sex Information, ethnic information, occupational information etc..The method, when realizing, firstly generates one and includes face and limbs Basis synthesis virtual image；Then particular emotion is selected；Then according to selected particular emotion, from virtual image Organ data storehouse selects and the face parts corresponding to this expression, substitute the corresponding face of described synthesis virtual image Parts, select the limbs organ corresponding with this expression from virtual image organ data storehouse, substitute described synthesis virtual The corresponding limbs organ of image so that the particular emotion that the virtual image finally given is selected before meeting.

Generating for virtual image of the present invention specifies the method for expression to be with a wide range of applications.Such as, different use Virtual image expression personal letter can be sent between family.In a preferred embodiment, see Figure 14, send out the use of personal letter Family A expresses one's feelings (" excited ") with drop-down menu setecting, as the second information of step S104, is then issued by personal letter User B.When user B opens this personal letter, first receive the first information i.e. virtual image of user A and describe information, First seen by step S101-S103 and not yet do the user's A virtual image expressed one's feelings；User B receives above-mentioned second By step S104-106, information, sees that user's A virtual image makes the expression (" excited ") specified by the second information.

In another embodiment, using the present invention's to generate, for virtual image, the method specifying expression, user can create Make and share the virtual image representing good friend or famous person.Such as, user C selects organ unit construction for representing good friend Or the virtual image of famous person D, and by this virtual image, as the first information, be sent on social networks, by User E receives this first information, and user E is it may be that but be not necessarily D；Second information is about this virtual image The expression to be shown, can be specified by above-mentioned user C or user E, it is also possible to specified by system；Pass through step S101-S106, is shown to user E, represents the virtual image with this appointment expression of user D.

Invention further provides and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, the method controlling the expression of user's virtual image, described virtual image is the virtual image that independent assortment generates, The implementation of its display expression is it may be that but be not necessarily, by step S101-S106；The method, based on one The parameter that individual or multiple applied environment residing for virtual image is relevant, is set as user by the expression meeting trigger condition The expression of virtual image.

In one embodiment, the emotional state of the independent assortment virtual image of band expression is associated with ambient parameter, Ambient parameter changes can trigger virtual image expression change.The associating policy of ambient parameter and virtual image expression is Can arrange；Such as, the expression strategy setting of himself is to obtain a bad board in gaming by a user Time, his virtual image shows the expression of " grieved "；And another user is by the expression strategy setting of herself For, the when of obtaining a bad board in gaming, her virtual image shows the expression of " angry ".Many arranging After planting the expression associating policy of ambient parameter and virtual image, the virtual image of band expression just becomes one can be anti- Reflect environment to change, change expression by user's personality, represent the robot of user.In another embodiment, User can also arrange the virtual image expression strategy of other users that he is seen, such as makes him see, other are used The virtual image at family obtains always showing the when of bad board the expression of " wailing " in gaming.

As a kind of preferred implementation, provide the user the expression strategy of a set of acquiescence and such as express one's feelings trigger condition list, With the interface that a user can customize expression strategy.In more detail, tactile by arranging different trigger conditions and its Send out device, generate or adjust expression strategy；Under the effect of expression strategy, when virtual image local environment is reached specific Trigger condition after, its trigger makes the virtual image of user show the expression specified by this trigger condition.At one In embodiment, it is also possible to automatically show the expression of association according to the application of user or game operation or other behavior, should Expression includes but not limited to that facial expression, gesture, posture, modal particle, sound, animation are play, sent band expression Message.The method of the expression of above-mentioned control user's virtual image acts on the virtual image that independent assortment generates, its The implementation of display expression is it may be that but be not necessarily, by step S101-S106.

Present invention also offers a kind of expression instrument, this expression instrument is implemented at least one and contains processor, memorizer, leads to The calculating device of news platform interface, at least one of which device also comprises localizer or touch screen, controls it for user The method of the expression of independent assortment virtual image, described virtual image is the virtual image that independent assortment generates, and it shows Show the implementation of expression it may be that but be not necessarily, by step S101-S106.The method comprises the following steps:

Step 1) receive and specify expression sequence, described expression sequence is an expression, or user is by slip localizer Or streak multiple expressions that touch screen selects；

Step 2) expression meeting trigger condition in above-mentioned expression sequence is set as the expression of user's virtual image.

Figure 10 is the schematic diagram of aforementioned expression instrument.In one embodiment, as it can be seen, the central part of this expression instrument Demultiplexing, in the virtual image of the band expression of Dynamic Display user, including expression mark, institute around core Stating expression to be designated user and provide a personal-machine interface, user can be by being operable to input table to expression mark Feelings (click on and express one's feelings mark or all can be considered that identifying input by expression expresses one's feelings expressing one's feelings to slip in mark, wherein, sliding Cross the virtual image expression previewing module that expression mark can be called in expression instrument, click on expression mark and can call expression instrument In virtual image expression handover module, by slip over continuously expression mark can produce expression sequence), then, table Feelings instrument process in accordance with the present invention S101-S106, or the implementation according to other display expression, at expression instrument Core shows the above-mentioned independent assortment virtual image with this expression.In one embodiment, except above-mentioned functions Outward, the expression that this expression instrument also can be inputted according to user exports corresponding modal particle, as with " happily " this table Modal particle corresponding to feelings " is laughed a great ho-ho ", and this modal particle realizes with word or with sound broadcasting.Additionally, in expression instrument Also can other integration of user interaction functionality such as integrated application or game operation.

In one preferably expression instrument embodiment, user is by slip localizer or streaks touch screen and selects a sequence Expression mark, the expression wherein meeting trigger condition becomes this expression shown by user's virtual image.Citing, on State trigger condition it may be that (such as unclamp mouse mouse-up, finger leaves when stopping localizer or touch screen operation Touch screen) localizer or finger be positioned at certain expression mark；Can also be that localizer or finger rest on certain table The feelings mark long enough time；Can also be that other user is mutual, such as double click or touch control gesture etc..At another In embodiment, above-mentioned expression sequence can trigger multiple expression to be shown, arranges display in order, as " laugh at-cry- Laugh at-cry-laugh at ", its implication is " not knowing whether to laugh or cry " (between tears and laughter).Above-mentioned expression instrument acts on certainly By the virtual image of combination producing, the implementation of its display expression it may be that but be not necessarily, pass through step S101-S106。

Present invention also offers and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, the method searching user with virtual image, described virtual image is the virtual image that independent assortment generates, Including:

Avatar information is specified in step 1, reception

Deposit due between virtual image and user that the aforementioned method expressed one's feelings by virtual image generation appointment of employing is generated At corresponding relation, therefore, it is possible to search user by known virtual image in turn.In one embodiment, User can search other users with virtual image, such as finds friend, or looks like certain star, or Meet oneself desired people in the heart.This embodiment is that customer group sets up a reverse indexing, from virtual image organ portion Part can obtain the user list using this organ parts as virtual image.Because it is one that the identity of virtual image describes The combination of individual virtual image organ parts, by the metadata of destination virtual image all parts and virtual image organ number Do according to the metadata in storehouse after mating, the user list meeting the organ parts of matching condition in data base is done in the lump Common factor processes, and can find the user meeting this virtual image condition.

In embodiment before, described virtual image organ data storehouse is existing data base.Real at another Execute in example, not there is ready-made virtual image organ data storehouse, or need to expand existing virtual image organ number According to storehouse, therefore the virtual image implementation method of the band espressiove of the present invention before step S101 also include foundation or Expand the step in virtual image organ data storehouse.

Three-dimensional avatars and the Different treatments of two-dimensional virtual image, essentially consist in, and 3D shape of face model is according to device Official's nominal contour hollows out, and each 3D organ model meets the type organ nominal contour, it is ensured that can be implanted to 3D In shape of face model, and (seeing Fig. 7 and Fig. 8) can be exchanged with same type 3D organ model.Two-dimensional virtual shape The Viewing-angle information of elephant by three dimensional structure's nominal contour information with substitute；Described by positional information is 3D model, and it is right Site information is relevant to three dimensional orientation, it is ensured that the correct of 3D organ model is implanted or replace；Expression information, body Part information, various attribute informations etc. then from two dimension the most substantially different.

Below, as a example by two-dimensional virtual image, describe the method set up or expand virtual image organ data storehouse in detail. With reference to Fig. 9, the step in described foundation or expansion virtual image organ data storehouse includes:

Step S401, obtain band espressiove virtual image original image, and mark Viewing-angle information, identity information, Expression information, positional information and classification information；

In the present embodiment, acquired virtual image original image is that art designing personnel are on the computing device by drawing The electronic format image file that software is drawn according to specification.By imaging or can also take the photograph in other embodiments Shadow model catches virtual image original image, and preserves with any file format.

In order to allow same type face organ, but the image of different virtual image, different expression can alternatively make With, art designing personnel need when drawing virtual image original image to ensure that image face visual angle is consistent with scaling. As a kind of preferred implementation, as it is shown in figure 1, the different virtual images of different expression all by one with nose The specific perspective cross square box of son-outspoken line, left eye-right eye straight line, determines its face contour and each face's organ Visual angle, size and location, so that it is determined that the visual angle of whole virtual image face and scaling.Draw these figures During sheet the face of fixing (relative to art designing personnel) and unified virtual image model towards, with based on model's face Size adjusts distance, thus ensures that its transparent effect is consistent with above-mentioned specific perspective cross square box, is drawn virtual shape The visual angle of elephant and organ size, position all meet specific criteria, just can make from different virtual images or show not Can alternatively use with the same type organ of expression, for combination producing is different but rational face.Said method is also Can be used for imaging or photographing, when catching model's human face expression, determine that its transparent effect is consistent with specific perspective cross square box, Thus ensure the replaceability of face's organ, for combination producing is different but rational face.

Each virtual image original image obtained by this step, it should mark its classification information, i.e. its institute comprised There is organ type；Its Viewing-angle information should be marked, such as generate the description letter of specific perspective cross square box used by image Breath；Should mark its positional information, example has an X-rayed the cross point of cross as described above as loci, marks this para-position Point position (generally, this loci is with the distance in the image upper left corner) in image or outside image boundary.If Needing, positional information can also comprise the information such as zoom scale, the anglec of rotation.If there is a default location, The original image meeting this default location can be without positional information.If there is an acquiescence visual angle, meet this and write from memory The original image recognizing visual angle can be without Viewing-angle information.If there is a default category, meet this default category Original image can be without classification information.

As a kind of preferred implementation, in the present embodiment, acquired virtual image original image be one or The multi-layer image file that one group can be alignd, hair, eyebrow, eye, nose, mouth, Er Deng face organ and skin, hands The limbs organs such as arm, thigh, trunk are positioned on different figure layers, when All Layers is shown in the lump between each organ Relative bearing correct.The related data of each figure layer includes and is positioned at face's organ of this figure layer or limbs organ Shape information, and classification information (if there being default category information, such as certain figure layer is defaulted as certain corresponding device Official's type, it is convenient to omit), and this figure layer is relative to the offset distance (if being not zero) of full images.Each figure Layer can be independently processed.

Original image set acquired in this step can have multiple virtual image, the original graph of each virtual image As the identity information of this virtual image should be marked to distinguish；Different virtual images can have respective multiple table Feelings, as smiled, dejected, cry, shy, startled, grieved etc., the original image of each band espressiove should be marked Note its expression information to distinguish.Virtual image these expression with eye, eyebrow, nose, the shape of Kou Deng face organ, In face, the position at place is relevant, also relevant (such as excitement with the attitude of the limbs organ such as arm, thigh, action Time, the attitude of arm or action can reflect this expression).This step should when obtaining virtual image original image Gather the most various expression, be beneficial to the generation of the combination virtual image of follow-up band espressiove.

Step S402, the virtual image original image obtaining step S401 are processed, and therefrom extract face's device Official or the graphic file of local face, and mark Viewing-angle information, identity information, expression information, positional information and class Other information.

In the present embodiment, from the virtual image original image of above-mentioned multi-layer image file, extract indivedual figure layer as face Portion's organ or the graphic file of local face (being referred to as face parts), and it is (original for this to mark its Viewing-angle information Image aspects information), identity information (for this original image identity information), expression information is (for this original image table Feelings information), positional information (for this original image location information, adjust by the offset distance of this figure layer) and class Other information (for the classification information of this figure layer).The information that there is default value can be omitted.

In other embodiments, the original image of organ figure layer is not divided to need through process to extract indivedual face parts, This process can be realized by face component model and organ contours shape recognition, such as identifies that nose is then by nose Region outside son cuts or becomes blank.

As a kind of optional implementation, from virtual image original image, extract face's organ or local face (system Be referred to as face parts) graphic file time, it is contemplated that face parts only account for it in whole virtual image original image In a part, therefore cut (or otherwise remove) blank parts around when extracting, only retain and be somebody's turn to do The picture of face parts boundary rectangle size is as graphic file, and calculates according to the above-mentioned information that cuts and mark para-position Vertex type and this loci position in picture or outside picture boundary, as positional information, show for correct This picture.If it is required, positional information can also comprise the information such as zoom scale, the anglec of rotation, according to above-mentioned sanction Information of cutting calculates.As a kind of optional implementation, when the virtual image original image gathered is multi-layer image file Time, i.e. hair, eyebrow, eye, nose, mouth, Er Deng face organ is positioned on different figure layers, by hair, eyebrow, eye, The figure layer at human face each places such as nose, mouth, ear intercepts out from virtual image original image respectively, then joins Blank parts around is removed according to said method.

As a kind of optional implementation, the face component diagram shape file extracted is not individually created or preserves, and It is otherwise to point to image resource, such as provides identification information specific image file, wherein map data mining platform, again Wherein cut information.

Described local face is the combination of several face's organs, therefore the local extraction of face graphic file and face The extraction of organ graphic file does not has difference substantially.Obtained one or more face component diagram shape file bags Include the shape information of described face parts, classification information (one or more organ type), identity information ( Or multiple virtual image identity) etc., additionally, one also should be included for the identification information distinguished.

Step S403, for each classification face's organ or local face graphic file do normalized, itself and more Its positional information new.

Following as a kind of implementation, face parts extraction obtained are processed respectively by organ type；Locally face The type in hole is the set of its composition organ type.Citing, is classified as a class by the graphic file of " mouth " and is uniformly processed. As a kind of optimal way, the minimum common boundary rectangle of same category of all face parts pictures can be taken as returning Category face component diagram chip size after one change, generates the generic face component diagram shape file being easy to replace.? When doing normalized, it should be ensured that the positional information of picture is correctly applied, calculates and marked.Further, The different expression graphic files of same category face parts can also be done at optimization the most respectively in additional examples Reason, for particular type and expression obtain may be less boundary rectangle.

As a kind of optional implementation, the face component diagram shape file after normalization is not individually created or preserves, But otherwise point to image resource, be such as given identification information specific image file, wherein map data mining platform, Wherein cut information.Citing, multiple generic face component diagram shape files merge into a file.

Step S404, for the face's organ in each classification or local face adjustment information or mark extraneous information, bag Include shape information, classification information, expression information, identity information, Viewing-angle information etc..

As a kind of optimal way, the face component diagram shape file of some band expression can be used to show multiple expression, Such as nose is not the most expressed one's feelings because making and is had huge change；Mark especially can be made for this type of graphic file, Or omit its expression information.

Optionally, may also include in this step and face component diagram shape file marked one or more attribute informations, Include but not limited to identification information, colouring information, age information, gender information, ethnic information, occupational information.

In one embodiment, the information of face parts can change the range of application of these face parts by adjusting, Such as adjust expression information, thus allow identity information or other attribute information qualified face parts for showing Show the expression outside the expression information that step S402-S403 is marked；One example of above-mentioned condition is facial muscle Organize by damage so expression shows abnormal special identity；Another example is to make or to wish to making spy The performer of different face or user identity.

In another embodiment, the image of face parts and relevant information can also change this face by calculating The attributes such as the shape of parts, color, the colour of skin；One example is " opening eyes wide ", the width of eye image constant but Highly it is multiplied by a multiplying power (such as 1.2 times)；Another example is " painting lip pomade ", adjusts the color of lip；Another Example is, changes its Visual Angle in Perspective by the affine transformation (affine transformation) of image.

Face's organ obtained by before step S405, storage or the information of local face, generate virtual image organ Data base.

In one embodiment, the information bag stored in organ data storehouse by face's organ or local face Including image (i.e. shape information) and metadata, this metadata can include that positional information, expression information, identity are believed Breath, Viewing-angle information, classification information, identification information, colouring information, age information, gender information, ethnic information, Occupational information etc..As a kind of preferred implementation, above-mentioned information can be by the shape information of face parts when storage (i.e. graphic file) is stored separately in a data acquisition system, and other information of face parts are stored in another In data acquisition system, can be associated by identification information between the data in two data acquisition systems.The advantage of do so It is easy to do data query, index, update, transmit and store.

More than complete to set up the detailed description of organ data base method.Do on the basis of already present organ data storehouse and expand Filling, it is possible to use essentially the same step, acquisition meets particular emotion information, identity information, Viewing-angle information etc. and wants The virtual image original image asked, processes and extracts the graphic file of face parts, optional normalized, mark Note relevant information, storage information the organ data storehouse after generating expansion.

During use, when classification information or other attribute information meet the inquiry by application settings or screening conditions, these Face image of component can be that application provides the candidate that can be combined as virtual image.When identity information is identical or Person matches, and when two dimensional image Viewing-angle information or 3-D view profile information are identical or match, works as expression information Meet the expression required for application, these face image of component can be used to combination producing application required for table The virtual image of feelings.

In one embodiment, art designing personnel draw 6 male of same view angle and 6 female according to step S401 The face original image of property virtual image, the eyebrow of these 12 virtual images, eyes, nose, mouth and shape of face are mutual The most different, the most only eyebrow, eyes, mouth has different shapes for different expressions, and each virtual image is each There are 18 kinds of expressions, altogether the face original image of 216 band espressioves.The organ of masculinity and femininity can not exchange, But the same type organ between male's virtual image, between women virtual image can exchange, because visual angle keeps Unanimously.Then, organ data base is generated according to remaining step.Do not consider that other virtual image appearance attribute is such as sent out Type, the colour of skin etc., this data base can combination producing 2 × 6 × 6 × 6 × 6 × 6=15552 different men, female's combination Virtual image identity, 5 kinds of organs of the most each combination virtual image identity be respectively from the same sex other 6 former Beginning virtual image, and each combination virtual image identity is respectively arranged with 18 kinds of expressions.Above-mentioned virtual image organ data storehouse, Therefrom select represent the virtual image of oneself and enable its virtual image to show this 18 kinds of tables for social network user Feelings.

Corresponding with said method, present invention also offers one and be implemented at least one calculating device, described meter Calculate device and include at least a processor, memorizer and communication platform interface, generate for independent assortment virtual image and refer to The system of fixed expression, described virtual image include but be not limited to two dimension or the personage of three-dimensional, animal, casper, Or abstract emoticon face, face's organ or local face acquired in this system come from one and are available for combination life Becoming the set of different virtual image, this system includes:

Present invention also offers one to be implemented at least one and calculate device, described calculating device is including at least at one Reason device, memorizer, set up or be extended for independent assortment virtual image and generate the virtual image organ data storehouse of expression System, including:

Original image acquisition module, obtains the virtual image original image of band espressiove, and marks Viewing-angle information, body Part information, expression information, positional information and classification information；

Parts extraction module, is processed virtual virtual image original image, therefrom extract face's organ or The locally graphic file of face, and mark Viewing-angle information, identity information, expression information, positional information and classification letter Breath；

Normalized module, for each classification face's organ or local face graphic file do normalized, And update its positional information；

Adjust or markup information module, adjust for the face's organ in each classification or local face or the extra letter of mark Breath, including shape information, classification information, expression information, identity information, Viewing-angle information etc.；

Image and metadata generation module, face's organ obtained before storage or the information of local face, generate Virtual image organ data storehouse.

Invention further provides and a kind of be implemented at least one containing processor, memorizer, the calculating of communication platform interface Device, controls the system of the expression of user's virtual image, and described virtual image is the virtual image that independent assortment generates,

Virtual image receiver module, receives and specifies avatar information；

The present invention has adaptability widely, both as shown in Figure 11 virtual image can be generated system (i.e. Noted above generates the system specifying expression for independent assortment virtual image) as social data server or trip The back-end system of play server, it is also possible to the system that generated by virtual image as shown in Figure 12 is as independent clothes Business supplier, it is also possible to the system that generated by virtual image as shown in Figure 13 creates system with virtual image organ Respectively as independent ISP.

Finally, it is to be noted that various technology described herein can be with combined with hardware or software, or in due course with two The combination of person realizes.Therefore, the method for presently disclosed subject matter, computer-readable medium and system or Its particular aspects or part can take to be included in such as floppy disk, CD-ROM, hard disk drive or other machine any The form of the program code (that is, instruction) in the tangible mediums such as readable storage medium storing program for executing, is wherein loaded when program code In the machines such as such as computer and when being executed by, this machine becomes the device for realizing this theme.

In the case of program code performs on programmable computers, calculating equipment generally can include processor, The readable storage medium (including volatibility and non-volatile memorizer and/or memory element) of this processor, at least One input equipment and at least one outut device.Can be such as by using data process API etc. to utilize this One or more programs of the establishment of the special programming model in territory of invention and/or each side of realization are preferably with senior Procedural language or OO programming language realize with computer system communication.If it is required, this program can To realize by assembler language or machine language.In any case, language can be compiler language or interpretative code, And realize combining with hardware.

It should be noted last that, above example is only in order to illustrate technical scheme and unrestricted.Although With reference to embodiment, the present invention is described in detail, it will be understood by those within the art that, to the present invention Technical scheme modify or equivalent, without departure from the spirit and scope of technical solution of the present invention, it is equal Should contain in the middle of scope of the presently claimed invention.

Claims

1. being implemented at least one and calculate a device, described calculating device includes at least processor, a memorizer With communication platform interface, generating the method specifying expression for virtual image, described virtual image includes but is not limited to two Dimension or the personage of three-dimensional, animal, casper or abstract emoticon face, the face's device acquired in the method Official or local face come from one for combination producing difference virtual image to represent the set of multiple user, the method Comprise the following steps:

The method specifying expression is generated the most according to claim 1, it is characterised in that described for virtual image One information include following at least one, the basis synthesis faces needed for specifying from multiple default basis synthesis faces Selection information, or one or more attribute information relevant with virtual image.

The method specifying expression is generated the most according to claim 2, it is characterised in that described for virtual image One or more relevant with virtual image attribute information included by one information, including following at least one, sex, Age, race, occupation.

The most according to claim 1 for virtual image generate specify expression method, it is characterised in that described in obtain The face's organ or the local face that take have related metadata.

The most according to claim 4 for virtual image generate specify expression method, it is characterised in that described with Face's organ or local face association metadata include following at least one: expression information, identity information, visual angle Information, profile information, positional information.

The most according to claim 4 for virtual image generate specify expression method, it is characterised in that described in obtain The face's organ taken or local face, the metadata being associated based on it.

The method specifying expression is generated the most according to claim 1, it is characterised in that described for virtual image Two information include following at least one:

The information that can act on virtual image sent from social networks.

The method specifying expression is generated the most according to claim 1, it is characterised in that also include for virtual image Storage can show independent assortment virtual image or the band Expression synthesis face specifying expression, and this storage information should for other With use.

The method specifying expression is generated the most according to claim 8, it is characterised in that by described for virtual image Virtual image band Expression synthesis face renders and is stored as picture, and described picture is static images or with expression animation Dynamic picture.

The method specifying expression is generated the most according to claim 8, it is characterised in that by institute for virtual image State virtual image band Expression synthesis face and be stored as reappearing the data message of this virtual image, described data message Including face's organ used by virtual image or the identification information of local face.

11. generate, for virtual image, the method specifying expression according to claim 1, it is characterised in that described Virtual image also includes that limbs organ, the limb action of described virtual image meet described appointment and express one's feelings.

12. generate, for virtual image, the method specifying expression according to claim 1, it is characterised in that described Collection is combined into virtual image organ data storehouse.

13. according to generating the method specifying expression described in claim 12 for virtual image, it is characterised in that also wraps Include the step set up or expand virtual image organ data storehouse.

14. generate, for virtual image, the method specifying expression according to claim 1, it is characterised in that user Include in the information sent to other users doing the virtual image expressed one's feelings.

15. generate, for virtual image, the method specifying expression according to claim 1, it is characterised in that user Make for other users and can do the virtual image expressed one's feelings, and this virtual image is sent on social networks as dividing Enjoy.

16. generate, for virtual image, the method specifying expression according to claim 1, it is characterised in that in step The second information received in rapid 4 is an expression sequence information, generates for virtual image in step 6 and shows this table The synthesis face of expression in feelings sequence.

17. 1 kinds are implemented at least one and calculate device, and described calculating device includes at least a processor and storage Device, the method setting up or expanding the virtual image organ data storehouse that band is expressed one's feelings, this data base is empty for combination producing difference Intending image and represent multiple user, the method step includes:

18. according to the method set up described in claim 17 or expand virtual image organ data storehouse, it is characterised in that The metadata of described face organ or local face has identity information, and described identity information is former based on its ingredient Virtual image belonging to beginning image.

19. according to the method set up described in claim 17 or expand virtual image organ data storehouse, it is characterised in that Described face organ or local face metadata include following at least one:

20. according to the method set up described in claim 17 or expand virtual image organ data storehouse, it is characterised in that Described face organ or local face metadata also include positional information, described positional information for this face's organ or Locally face location in whole virtual image.

21. according to the method set up described in claim 17 or expand virtual image organ data storehouse, it is characterised in that Also include for described face organ or local face mark at least one attribute information, described attribute information include but not It is limited to identification information, classification information, colouring information, age information, gender information, ethnic information, occupational information.

22. 1 kinds are implemented at least one containing processor, memorizer, the calculating device of communication platform interface, control The method of the expression of user's virtual image, described virtual image is the virtual image that independent assortment generates, and its feature exists In:

23. 1 kinds are implemented at least one and contain processor, memorizer, the calculating device of communication platform interface, wherein At least one device also comprises localizer or touch screen, the method controlling the expression of user's virtual image, described virtual Image is the virtual image that independent assortment generates, including:

24. 1 kinds are implemented at least one and contain processor, memorizer, the calculating device of communication platform interface, with void Intending the method that image searches user, described virtual image is the virtual image that independent assortment generates, including:

Avatar information is specified in step 1, reception；

25. 1 kinds are implemented at least one and calculate device, and described calculating device includes at least processor, a storage Device and communication platform interface, generate the system specifying expression for virtual image, and described virtual image includes but is not limited to Two-dimentional or three-dimensional personage, animal, casper or abstract emoticon face, the face acquired in this system Organ or local face come from one for combination producing difference virtual image to represent the set of multiple user, and this is System includes:

26. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State the first information include following at least one, from multiple default basis synthesis faces specify needed for basis synthesis faces The selection information in hole, or one or more attribute information relevant with virtual image.

27. according to claim 26 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State one or more relevant with the virtual image attribute information included by the first information, including following at least one, Sex, age, race, occupation.

28. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute Face's organ or the local face of stating acquisition have related metadata.

29. according to claim 28 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State include with face's organ or the local metadata that associates of face following at least one: expression information, identity information, Viewing-angle information, profile information, positional information.

30. according to claim 28 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State face's organ or the local metadata that is associated based on it of face of acquisition.

31. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute Second information of stating include following at least one:

The information that can act on virtual image sent from social networks.

32. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that also Including a memory module, for storing the independent assortment virtual image or band Expression synthesis face that can show appointment expression, Other application need not described confession combination producing difference virtual image to represent multiple user's when using this storage information Set.

33. according to claim 32 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute Stating memory module and render and be stored as picture by described virtual image band Expression synthesis face, described picture is static map Sheet or the dynamic picture with expression animation.

34. according to claim 32 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State memory module and described virtual image band Expression synthesis face is stored as reappearing the data letter of this virtual image Breath, described data message includes face's organ used by virtual image or the identification information of local face.

35. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State virtual image and also include that limbs organ, the limb action of described virtual image meet described appointment and express one's feelings.

36. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that institute State collection and be combined into virtual image organ data storehouse.

37. according to claim 36 generations for virtual image specify the system expressed one's feelings, it is characterised in that also Including the module setting up or expanding virtual image organ data storehouse.

38. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that use Family includes in the information sent to other users doing the virtual image expressed one's feelings.

39. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that use Family is that other users make and can do the virtual image expressed one's feelings, and this virtual image is sent on social networks as Share.

40. according to claim 25 generations for virtual image specify the system expressed one's feelings, it is characterised in that the The second information that two information receiving modules are received is an expression sequence information, and band Expression synthesis face generation module is Virtual image generates and shows the synthesis face of expression in this expression sequence.

41. 1 kinds are implemented at least one and calculate device, and described calculating device includes at least a processor and storage Device, sets up or expands the system in the virtual image organ data storehouse that band is expressed one's feelings, and this data base is empty for combination producing difference Intending image and represent multiple user, this system includes:

42. foundation according to claim 41 or the system in expansion virtual image organ data storehouse, its feature exists In, the metadata of described face organ or local face has identity information, and described identity information is based on its composition portion Divide virtual image belonging to original image.

43. foundation according to claim 41 or the system in expansion virtual image organ data storehouse, its feature exists In, described face organ or local face metadata include following at least one:

44. foundation according to claim 41 or the system in expansion virtual image organ data storehouse, its feature exists In, described face organ or local face metadata also include positional information, and described positional information is used for this face's device Official or face location in whole virtual image, local.

45. foundation according to claim 41 or the system in expansion virtual image organ data storehouse, its feature exists In, also including marking at least one attribute information for described face organ or local face, described attribute information includes But it is not limited to identification information, classification information, colouring information, age information, gender information, ethnic information, occupation Information.

46. 1 kinds are implemented at least one containing processor, memorizer, the calculating device of communication platform interface, control The system of the expression of user's virtual image, described virtual image is the virtual image that independent assortment generates, and its feature exists In, this system includes:

47. 1 kinds are implemented at least one and contain processor, memorizer, the calculating device of communication platform interface, wherein At least one device also comprises localizer or touch screen, controls the system of the expression of user's virtual image, described virtual Image is the virtual image that independent assortment generates, and this system includes:

48. 1 kinds are implemented at least one and contain processor, memorizer, the calculating device of communication platform interface, with void Intending image and search the system of user, described virtual image is the virtual image that independent assortment generates, and this system includes:

Virtual image receiver module, receives and specifies avatar information；