JP2007026088A - Model creation apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a model creation apparatus allowing even an amateur to easily create model data, and allowing a user to obtain a wide variety of model data by simply selecting desired parts, etc. <P>SOLUTION: The model creation apparatus 14 extracts feature points of a face image using a feature point extracting part 24 and causes a feature point display part 26 to display the feature points. The differences between the feature points and the face, etc., are corrected and the feature points shown by input model data (template data) 40 are changed into the feature points of the face image by a face deforming part 28 to deform an input model. Model data 42 deformed after the face image supplied, and desired parts data 56 selected by means of a parts setting part 30 are supplied to an image sticking part 32. A model image stuck according to both data 42 and 56 is drawn. In this way, the user can easily create basic model data 62 regardless of his or her knowledge of the model. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a model generation device. In particular, the present invention relates to an apparatus for generating an avatar model that expresses movements and expressions of humans and animals by computer graphics.

  Conventionally, there is a proposal in Patent Document 1 in which communication is performed using an avatar which is a virtual human model in which an image is pasted on a polygon. In the proposed method, first, for example, image data of the user or the other party who is a call target input from a camera is acquired. In this method, the movement of the call target is obtained from the next acquired image data, and facial expression information is detected based on the obtained movement. In the next stage, a drawing image is created by rendering a model created in advance according to the detected facial expression information. By making a call while looking at the created drawing image, this proposal realizes communication taking into consideration the privacy of the communication target while capturing the emotion of the call target in communication like a video phone.

  In order to develop a service using such a method, it is required to provide a variety of models to the user easily at low cost. However, specialized knowledge is required for model creation. Due to such demands, model creation is costly and there is a problem that a sufficient amount cannot be provided to the market.

  Conventionally, Patent Document 2 has been proposed for such a problem. This proposal is a method in which a user creates a model by selecting a face part.

Japanese Patent No. 3593067 Japanese Patent No. 2294703

  However, in the method of Patent Document 2, many parts that require knowledge of creating a three-dimensional model are created in advance, and the user selects them. For this reason, a great deal of cost and time are required in the stage of creating parts.

  The present invention eliminates the drawbacks of the prior art, and provides a model generation apparatus that can easily create model data even for an amateur and obtain a wide variety of model data only by a simple operation of selecting a desired part or the like by a user. The purpose is to do.

  In order to solve the above-mentioned problems, the present invention includes a face image used for characterizing each face in advance, part data for modifying the face image, and a first feature point representing a preset face part. In the model generation apparatus that inputs basic model data to be generated and generates new basic model data reflecting the three input data, the apparatus extracts a feature point in the face image as a second feature point A feature point extracting unit, a deforming unit that associates the first feature point with the second feature point, and moves the first feature point to the second feature point, and a part that selects a desired part from the part data The image processing apparatus includes: selection means; and pasting means for pasting an image based on the moved first feature point and the selected part data.

  The model generation apparatus of the present invention extracts the second feature point by the feature point extraction unit, displays the second feature point on the first display unit, corrects the deviation between the feature point and the face, and the like. The first feature point is changed to the second feature point by means to change the model data to resemble a face image, and the deformed model data and desired part data selected by the part selection means are pasted. The model image pasted on the basis of both data is drawn, and model data can be easily created regardless of the presence or absence of model knowledge.

  Next, an embodiment of a model generation apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

  In this embodiment, the model generation apparatus of the present invention is applied to the model generation system 10. The illustration and description of parts not directly related to the present invention are omitted. In the following description, the signal is indicated by the reference number of the connecting line in which it appears.

  As shown in FIG. 1, the model generation system 10 includes a data storage unit 12 and a model generation device 14. The data storage unit 12 has a function of storing various data. The data storage unit 12 is a face image storage unit 16 that stores face image data, input model data, part data, and output model data as various data, an input model storage unit 18, a part data storage unit 20, and an output model storage unit 22. including. The model generation system 10 sends an operation instruction from an operation unit (not shown) to the control unit. The control unit outputs a control signal generated according to the operation instruction to the data storage unit 12. The data storage unit 12 outputs data from each of the face image storage unit 16, the input model storage unit 18, and the part data storage unit 20 to the model generation device 14 in accordance with the supplied control signal.

  The face image storage unit 16 stores realistic human face image data taken by a camera or comic image data created by a portrait tool or the like. In this embodiment, the case where comic image data is stored is shown in the latter part. The comic image data is stored in combination with the transmission information. The face image data has, for example, 32 bits of information per pixel. Of this information, 24 bits represent RGB data, and the remaining 8 bits represent transmission information. The face image data is created by a user who operates the model generation device 12. The user here is, for example, a content provider.

  The input model storage unit 18 stores a template model. In the template model, a plurality of planar meshes are arranged slightly shifted in the depth direction. For the mesh, a face mesh, a part mesh, etc. are prepared. A standard human polygon is placed on the face mesh. Further, for example, meshes that increase variations of models assumed in advance, such as body meshes, eyeglass meshes, and beard meshes, are prepared as part meshes. Each part mesh is placed in the correct position in advance. For example, the eyeglass mesh is placed at the position of the eyes in the face mesh. The beard mesh is arranged at the mouth position of the face mesh.

  Furthermore, for example, if the face mesh moves to the left and right, the operation is set for each mesh so that the eyeglass mesh follows and moves. Here, an image to be pasted on the surface of the mesh is called a texture. The texture of each mesh is set to a transparent image. Feature points are set in advance on characteristic parts of the face such as the eyes, mouth, and nose. Since the feature point changes according to the face image, it is also called a change point.

  The part data storage unit 20 is part data suitable for the part mesh in the template model, and also has transparency information. The user creates part data.

  The model generation device 14 includes a feature point extraction unit 24, a feature point display unit 26, a face deformation unit 28, a part setting unit 30, an image pasting unit 32, and a model display unit 34. The feature point extraction unit 24 has a function of extracting feature points included in the face image data 36 supplied from the face image storage unit 16. The feature point extraction unit 24 outputs the feature point data 38 in the extracted face image to the feature point display unit 26 and the face deformation unit 28.

  The feature point display unit 26 has a function of inputting the supplied feature point data 38, displaying the feature point data 38 on the screen as the feature points of the face image, and editing the displayed feature points. The feature point display unit 26 outputs the feature point data 38 edited according to the instruction from the operation unit to the face deforming unit 28.

  The face deformation unit 28 has a function of moving the feature point of the input model selected according to the feature point data 38 as a change point. The face deformation unit 28 is supplied with input model data 40 that is a template model from the input model storage unit 18. The input model data 40 is selected according to an instruction signal supplied to the input model storage unit 18 via the operation unit. The face deforming unit 28 inputs the selected input model data 40 and changes the feature points of the input model data 40 according to the feature point data 38. The face deforming unit 28 can diversify the input model by changing the input model in accordance with each supplied face. The face deforming unit 28 outputs the deformed face model data 42 to the image pasting unit 32.

  The part setting unit 30 has a function of selecting and setting appropriate part image data from the part image data. The part setting unit 30 is supplied with planar mesh data 44 in the template model from the input data storage unit 18. The parts setting unit 30 is supplied with parts data 46 from the parts data storage unit 20. In order to realize the above-described function, the parts setting unit 30 includes a parts display unit 48, a parts determination unit 50, and a title input unit 52 as shown in FIG.

  The part display unit 48 has a function of displaying the supplied part data 46. The part display unit 48 displays the part data 46 as a part as will be described later, and supplies the part data 54 corresponding to the displayed part image to the part determination unit 50.

  The part determination unit 50 has a function of selecting and determining from a plurality of types of supplied part data. The parts determination unit 50 outputs the selected parts data 56 to the title input unit 52. The title input unit 52 has a function of assigning a title to each determined part. The title input unit 52 is supplied with data 44 from the input data storage unit 18.

  The parts determination unit 50 and the title input unit 52 correspond to a partial function in the operation unit, that is, a determination key function and a key input function.

  Returning to FIG. 1, the image pasting unit 32 creates the model data of the deformed face image and the selected part data as new basic model data, and further pastes the created basic model data as an image. Have The image pasting unit 32 receives the deformed face model data 42 including the face mesh from the face deforming unit 28, and sets the face image as the texture of the face mesh. At this time, as will be described later, for example, by setting the transparency information of the face image, the texture of the face mesh is set so that only the face area can be seen.

  The image pasting unit 32 receives the mesh of each part and the part data 58 from the part setting unit 30, and sets the selected part image as the texture of the part mesh. If the transparency information of the part image is set at this time, the texture of the part mesh is set so that, for example, only the frame portion of the glasses can be seen.

  The image pasting unit 32 arranges and renders the face mesh and the part mesh in the same manner as the template model. The rendered image data 60 is output to the model display unit 34. The image pasting unit 32 displays reference model data 62 that can integrally represent a template model indicating all meshes, a face texture, and a plurality of part textures when it is determined that the model displayed by the user is good or bad. Generate and output to the output model storage unit 22.

  The model display unit 34 has a function of displaying a model. The model display unit 34 is supplied with input model data 64 and image data 60 which are template data before model construction. The model display unit 34 can confirm the quality of the model by displaying both data 60 and 64. The model display unit 34 may be the same monitor as the feature point display unit 26. The same monitor may display corresponding data in accordance with display contents in the multi-screen.

  The content provider uses the model generation device 14 to create new reference model data 62 from the input model data 40 and mesh data 44, which are template models, and a general user uses the model generation device 14 to generate reference model data. Create personal model data from

  Next, the operation of the model generation system 10 will be described. First, a procedure for creating reference model data will be described.

  The feature point extraction unit 24 takes in the face image data 36 and extracts a face part of the face image. In this embodiment, the detection is automatically performed using a known technique disclosed in Patent Document 1. At this time, at least the change points represented by the vertices of the black triangle set in the template model of FIG. 3 are extracted. Here, the feature points are extracted with two-dimensional coordinates with the origin 66 at the upper left of the face image. However, feature point extraction may fail. The feature point display unit 26 may display the face image and the extracted feature points in an overlapping manner. When the content provider sees the display and the feature points are shifted, the extracted feature points can be corrected. A feature point can be added when not extracted.

  When the correct feature point is determined in this way, the face deforming unit 28 receives the information of all the feature points and the input model data 40, that is, the template model. In particular, the change point set in the template model and this change are received. Reference is made to the coordinates of the feature points corresponding to the points. With this reference, the corresponding change point is moved to the feature point. By this process, a face mesh 67a matching the input face image is created as shown in FIG. FIG. 4 shows not only the face mesh 67a but also a glasses mesh 67b, a beard mesh 67c, and a body mesh 67d.

  As shown in FIG. 5 (a), the image pasting unit 32 receives the face model data 42 including the face mesh and the face image from the face deforming unit 28, and sets the face image as the texture of the face mesh. At this time, the transmission information of the face image as shown in FIG. 5 (b) is set. The image pasting unit 32 is set so that only the face area can be seen as the texture of the face mesh.

  The part setting unit 30 reads a plurality of part data 44 and displays it as a part image. The content provider selects a part that matches the face image from the displayed part images. For example, when the face image is a woman as shown in FIG. 5 (a), for example, several part images of female glasses for glasses mesh shown in FIG. 6 (a) are set. Moreover, you may set several parts images of the clothes for women for a body mesh. However, the beard mesh is not set because it does not fit women.

  Each time one part image is determined in the part setting unit 30, the title input unit 52 displays a display window for inputting the title of the part image. Enter the appropriate title (text data). Then, the title is stored as information along with the part texture in the model. In the data (reference model data) 58 generated in this way, a title is assigned to each of a plurality of part textures.

  The image pasting unit 32 receives the mesh and part data of each part from the part setting unit 30, and sets the selected part image as the texture of the part mesh. At this time, as shown in FIG. 6 (b), when the transparency information of the part image is set, the texture of the part mesh is set so that, for example, only the frame portion of the glasses can be seen.

  The image pasting unit 32 arranges and renders various data supplied according to the above-described settings in the same manner as the template model. When the image data 60 obtained by this rendering is supplied to the monitor display unit 34, the content provider can check the generated model. The image pasting unit 32 generates new reference model data 62 in which a template model of all meshes, a face texture, and a plurality of part textures are integrated.

  The model generation system 10 is not limited to the generation of a cartoon image model of a face image, and a content provider can also create a model by an individual user with his / her face photograph. In this case, since the comic image is not included, the face mesh is in a so-called “Nopperabo” state. Therefore, the model generation device 14 does not perform feature point extraction or face deformation.

  A procedure for creating an individual model by an individual user will be described. For the face mesh, a cartoon image texture or a transparent texture like Nopperabo is set. The part mesh has a plurality of part textures for each part.

  When setting a transparent texture to the input model data, a real face image is used. In this setting, there is generally no transparency information. On the other hand, when a comic image is used, a face image is not used. Also, parts images are not used.

  Of the above-described conditions, when a face image is input as a real image, the feature point extraction unit 24 extracts a feature point and estimates a face region from the extracted feature point. From the estimation of the face area, the face area is determined. The feature point display unit 26 corrects the extracted feature points and face area.

  In this case, the face deformation unit 28 deforms the face mesh based on the feature point data 38. The deformed face model data 42 is output to the image pasting unit 32. In the image pasting unit 32, the determination part of the face area is made opaque and used as a face image texture. The image pasting unit 32 sets a region other than the determination portion as a transparent texture.

  If no face image is input, a cartoon texture is supplied to the face mesh. The feature point extraction unit 24, the feature point display unit 26, and the face deformation unit 28 do not perform any processing, and the image pasting unit 32 does not process the face mesh.

  The part setting unit 30 generates a part texture from each part data included in the model data, and displays each part as an image. The individual user selects one favorite part for each part from the displayed parts. The selected part texture 58 is supplied to the image pasting unit 32. In the image pasting unit 32, the image pasting unit 32 for pasting the part texture 58 supplied to the corresponding part region supplies the part texture 58 pasted on the model display unit 34 and the image data 60 obtained by rendering the face mesh. The model display unit 34 displays the image data 60.

  If the individual user determines that this model is acceptable, a personal model in which all mesh face image models, face textures, and selected part textures are integrated is generated as a personal basic model.

  Once the template model is created by operating in this way, reference model data with various faces and parts can be easily created from the created template model, and various personal basic models can be created from the created reference model data. Can create data.

  Further, in the process of creating the reference model data, the content provider need only design images such as caricatures and parts, so that complicated knowledge in model generation is not required. Furthermore, since the process of creating personal basic model data requires only terminal operations such as camera shooting and parts selection, it can be handled easily by ordinary users.

  As described above, the model generation system 10 can provide and develop a wide variety of basic model data to the market, thereby contributing to sales expansion of tools in communication communication using the provided model data.

  Also, the parts setting unit 30 is not limited to the configuration shown in FIG. 2, and may set parts based on information already input as shown in FIG. In this case, the parts setting unit 30 uses the parts display unit 48 and the parts determination unit 50 shown in FIG.

  The title reading unit 68 has a function of reading the title of each part texture existing in the input model storage unit 18. The title read unit 68 generates a read control signal 44 according to the input key 72 and outputs it to the input model storage unit 18. The input model storage unit 18 supplies the title 44 of the part texture read in response to the read control signal 44 to the title read unit 68. The title reading unit 68 outputs the supplied title 44 as the title 74 to the search unit 70.

  The search unit 70 has a function of collating the supplied title 74 with the input key 72 from the user and detecting a matching title. The search unit 70 outputs the title 76 matched by the detection function to the parts display unit 48.

  The part display unit 48 displays a part texture corresponding to the searched title 76 among the part textures in the supplied title 44. A plurality of part textures may be displayed. The parts display unit 48 outputs the displayed part texture 54 to the parts determination unit 50. The part determination unit 50 selects some of the displayed part textures. The part determining unit 50 outputs the selected part texture data 58 to the image pasting unit 32.

  The operation of the model generation system 10 is basically as described above. In particular, when the content provider operates using this configuration, the part display unit 48 displays a matched part image among the supplied part images 44. Next, the part determination unit 50 selects and determines a part that matches the face mesh from the displayed part images.

  Further, when creating personal model data, in the part setting unit 30, the title reading unit 68 passes the part texture title 44 read from the input model storage unit 18 to the search unit 70. Next, the search unit 70 collates the input key 72 supplied from the individual user with the title 74 and outputs the matched title 76 to the parts display unit 48.

  The parts display unit 48 receives the title 76 from the search unit 70 and displays only the part texture 44 having this title. Further, the parts determination unit 50 selects and determines a part user's preference from among the part textures displayed on the part display unit 48 according to the input data 72.

  When operated in this way, when the number of part textures becomes enormous, an individual user can search for a desired part with the input key, and thus can easily create a favorite model. Further, it is advantageous for use in a terminal device having a limited display area such as a mobile phone.

  Further, the model display unit 34 is not limited to the configuration shown in FIG. 2, and may flexibly correspond to the face image as shown in FIG. In this case, the model display unit 34 in FIG. 8 not only has a display function in the above-described configuration, but also includes a face analysis unit 78 and a model render unit 80. The face analysis unit 78 has a function of analyzing face image data 84 input from an external image input device, for example, a digital camera 82 and detecting feature points as facial expression information 86. The model renderer 80 has a function of rendering the supplied input model data 64 based on the facial expression information 86.

  When the model display unit 34 receives the image data 60 of the basic model on which image pasting has been performed in the previous processing, the model display unit 34 activates the camera 82. At this time, the face image of the subject is continuously taken into the face analysis unit 78 of the model display unit 34.

  The face analysis unit 78 receives continuously sent face images and detects a characteristic portion or part indicated by the face image data as, for example, a feature point. The movement of the face, that is, facial expression information, is created from the change of the feature point in the face image continuously supplied in time series by the detection of the feature point. The model renderer 80 uses the facial expression information 86 sequentially supplied from the face analyzer 78 to display the model while deforming it. This process is called face sync. An example of the model displayed here is shown in FIG. Here, the creation of facial expression information and details of each of the face syncs are disclosed in Patent Document 1. The present embodiment uses this disclosed technique.

  When the model display unit 34 is configured and operated in this manner, the texture of the face image and the texture of the part image are selected, and when the model state is confirmed after setting, the created model is actually You can see how it moves. This is advantageous in that it can be verified before actually applying whether or not a model as intended by the individual user has been created.

  Next, a model providing system 90 to which the model generation system according to the present invention is applied will be described. The model providing system 90 is a distributed system in which at least the components of the model generation system 10 are organically connected by communication. As shown in FIG. 10, the model providing system 90 includes server devices 92 and 94 and a user terminal device 96 as components.

  The server device 92 has a function of creating and storing input model data, that is, a template model, and outputting the template model in response to access. The server device 92 includes the data storage unit 12 shown in FIG. The server device 92 of this embodiment is a case where a template model already created by a creator who has specialized knowledge of model creation is stored. The template model stores not only human template models but also various denplate models such as dogs and cats. Therefore, the storage included in the server device 92 corresponds to the input model storage unit 18 of FIG. Server device 92 is connected to network 98.

  The server device 94 has a function of creating the reference model 62 from the template model, and is connected to the network 98. The network 98 may be an IP (Internet Protocol) network. In order to realize this function, the server device 94 includes a model generation device 14. The server device 94 reads the template model from the server device 92 and downloads it via the network 98. The server device 94 is managed by a distributor who distributes content, for example, a content provider. This management includes payment of a usage fee to the content provider, that is, a billing process. The server device 94 includes the output model storage unit 22 of FIG.

  The user terminal device 96 is a device owned by an individual user. User terminal device 96 is also connected to network 98. The user terminal device 96 corresponds to the input operation unit, and determines whether the created new reference model data 62 is acceptable.

  The operation of the model providing system 90 will be briefly described. The user terminal device 96 creates specification information required for creating a desired reference model data, particularly a personal model such as a part texture and a face image. The user terminal device 96 accesses the server device 94 from a personal computer or a mobile phone, for example, and transmits the created specification information to the server device 94 as order information.

  The server device 94 reads input model data corresponding to the specification information from the server device 92 through user management based on the presence / absence of user registration and billing management for each user. The server device 94 supplies the input model data to the model generation device 14, and creates reference model data and personal reference model data by applying the method described above. Needless to say, various reference model data 62 is created for each facial image. As human model data, there are endless variations such as young women, elderly men, children, and babies. The server device 94 is also a website that introduces the created reference model data to general users, that is, users who use the reference model data as a communication tool. The website accepts orders from individual users as described above. As described above, the server device 94 operates the website by user management, individual user management, and billing management. The content provider may settle the fee at the time of downloading the model data, or may collect a flat fee from the user every month by a membership registration system.

  The user terminal device 96 downloads a personal reference model based on its own order information created by the server device 94 via the network 98. The user uses the personal reference model as a communication tool of the user terminal device 96.

  According to the model providing system 90, the content provider can easily create a variety of reference models simply by designing a two-dimensional image such as a cartoon image or a part image and provide it to the user. From this, it becomes possible to operate the site in the system 90 so that the user is not bored.

  In addition, users can select desired model data from abundant reference models, and acquire unique model data for personal use, such as the ability to make face images into photographic images or select parts freely. Can do.

  In this way, when the system 90 is implemented, the content shortage that has been a problem can be solved. For this reason, it is possible to contribute to sales expansion of communication tools. Further, in this embodiment, since the model data is generated by the server device 94, there is an advantage that it is easy to use for a mobile phone user without imposing a burden on the user terminal device.

  Note that the user terminal device 96 may include a model generation device 14 as shown in FIG. The user terminal device 96 downloads application software used by the model generation device from the server device 94 managed by the content provider. Next, the user terminal device 96 downloads desired reference model data from the reference model created by the content provider from the server device 94. The user terminal device 96 is a built-in model generation device 14, and creates personal reference model data from reference model data. Accordingly, the user terminal device 14 uses the generated personal reference model with a communication tool or the like.

  In this way, in the model providing system 90, the generation of personal reference model data is processed by each user terminal device 96, whereby the server device 94 eliminates the need for detailed order exchanges with many users and suppresses the network load. The effect to do. This allows content providers to operate sites for more customers at lower cost.

1 is a block diagram illustrating a schematic configuration of a model generation system to which a model generation apparatus according to the present invention is applied. It is a block diagram which shows the schematic structure of the parts setting part shown in FIG. It is an example of the input model data (template model) stored in the input model memory | storage part of FIG. It is a figure explaining each mesh in the template model which the model generation system of FIG. 1 handles. It is a figure explaining the face picture collection which the model generation system of FIG. 1 handles, and its transparency information. It is a figure explaining the parts picture collection which the model generation system of FIG. 1 handles, and its transparency information. It is a block diagram which shows the schematic structure different from the parts setting part of FIG. It is a block diagram which shows arrangement | positioning of the component added in the model display part 34 of FIG. It is a figure which shows the example of a display of the face sync obtained with the model generation system of FIG. It is a block diagram which shows the schematic structure of the model provision system to which the model generation system which concerns on this invention is applied. FIG. 11 is a block diagram showing components added in the user terminal device of FIG.

Explanation of symbols

10 Model generation system
12 Data storage section
14 Model generator
16 Face picture storage
18 Input model storage
20 Parts data storage
22 Output model storage
24 Feature point extraction unit
26 Feature point display
28 Face transformation
30 Parts setting section
32 Image pasting part
34 Model display

Claims (7)

Basic model data including a face image used for characterizing each face in advance, part data for modifying the face image, and a first feature point representing a preset face part are input and input. In a model generation device that generates new basic model data reflecting three data, the device includes:
Feature point extracting means for extracting a feature point in the face image as a second feature point;
Deformation means for causing the first feature point to correspond to the second feature point and moving the first feature point to the second feature point;
Parts selection means for selecting a desired part from the parts data;
A model generating apparatus comprising: pasting means for pasting an image based on the moved first feature point and the selected part data. The apparatus according to claim 1, wherein the apparatus includes first display means for displaying a second feature point;
And a second display means for displaying a model image obtained by pasting. 3. The apparatus according to claim 1, wherein the part selection unit includes a third display unit that displays the part data as an image;
A determining means for selecting a desired part from the displayed part image;
An information input unit for adding additional information to part data corresponding to a selected part. The apparatus according to claim 1 or 2, wherein the part selection means reads out the additional information and the part data based on the additional information attached to the already created part data,
Search means for collating the additional information read out under the control of the reading control means with the input key information and detecting additional information matching the key information;
Third display means for displaying the part data corresponding to the detected additional information as an image;
A model generation apparatus comprising: a determination unit that selects a desired part from an image of displayed part data. 5. The apparatus according to claim 2, 3 or 4, wherein the second display means is a facial motion obtained based on a face image continuously supplied in time series from an image input means for digitally inputting a face image. Facial expression analysis means for detecting facial expression information,
A model generation apparatus including a drawing unit that draws a model image by pasting an image in which the feature points are changed using the facial expression information. The model generation system in which the model generation device according to any one of claims 1 to 4 is arranged, the system includes:
A first server device for storing the model data;
A second server device in which the model generation device is arranged;
Output order information related to the creation of the new basic model data, including a terminal device for acquiring the created basic model data,
The first and second server devices and the terminal device are connected to a communication network,
A model generation system characterized in that the new basic model data is acquired for personal use.   The model generation system according to claim 6, wherein the terminal device includes the model generation device.

Images