JP6697573B2 - Hairstyle simulation device, hairstyle simulation method, and computer program - Google Patents

Description

  The present invention relates to a hairstyle simulation device, a hairstyle simulation method, and a computer program that can simulate a hairstyle selected by a user by transforming the hairstyle into a desired shape.

  A large number of simulation systems using images have been developed in order to visually confirm whether or not a user has a hairstyle that suits him / herself. For example, Patent Document 1 discloses a beauty simulation system capable of simulating makeup and hairstyle and displaying the results after the simulation side by side.

  In addition, Patent Document 2 discloses a hair simulation device capable of adjusting a simulated hairstyle so as to have a natural finish. In Patent Document 2, it is possible to adjust the hairstyle image by setting the hairstyle image as an image of a plurality of layer structures and deforming the polygon for each layer structure.

JP, 2013-178789, A Japanese Patent No. 5422033

  However, in Patent Document 2, since the polygon is deformed for each layer structure, even if it is desired to deform only a part of the hairstyle image, the hairstyle image of the part belonging to the same layer structure is also deformed. Therefore, there is a possibility that a part of the hairstyle image that does not need adjustment may be deformed, and it may be difficult to adjust the hairstyle image to a desired shape as a whole.

  The present invention has been made in view of the above circumstances, and provides a hairstyle simulation device, a hairstyle simulation method, and a computer program capable of simulating a hairstyle selected by a user by transforming the hairstyle into a desired shape. The purpose is to do.

In order to achieve the above object, a hairstyle simulation device according to a first aspect of the present invention is a hairstyle simulation device capable of transforming only a part of a model image showing a hairstyle that is designated by a user. An image capturing unit that captures a face, a contour extracting unit that extracts a contour line of a face portion from the image data of the captured face, and a contour line of the extracted face portion are extracted from model image data indicating a hairstyle. By matching the contour line with the contour line of the face portion and the contour line, the face portion of the image data of the imaged face and the hair portion of the model image data are superposed to generate a composite image. And a composite image display means for displaying on a touch display, wherein the hair portion of the model image data is displayed as a polygon with texture represented by a node / link structure of a quadrangle primitive. Operation detecting means for detecting a predetermined operation on a part of the hair part of the composite image displayed on the display, and image deforming means for deforming the shape of the hair part according to the detected predetermined operation. And the operation detecting means detects a change area, a moving direction and a movement amount as the predetermined operation, and the image deforming means, based on the detected moving direction and the movement amount, for the detected change area. And a deformation calculation means for calculating a deformation direction and a deformation amount of a polygon with a texture represented by a node / link structure using a quadrilateral primitive, and the deformation calculation means provides an auxiliary point inside the quadrilateral primitive. A feature is that a textured polygon represented by a node-link structure of four triangular primitives having points as vertices is gradually deformed while interlocking with each other .

Further, hairstyle simulation apparatus according to the second invention, have you the first invention, the operation detecting means to detect a predetermined range from the coordinate position of the synthetic image displayed on the touch display as a change region Is preferred.

Also, in the hairstyle simulation device according to the third aspect of the present invention, in the first or second aspect of the present invention , the deformation calculation means calculates the maximum amount of deformation of the node forming each primitive from the contour line of the face portion of each node. Depending on the distance, it is preferable that the shorter the distance, the smaller the setting.

Next, in order to achieve the above object, the hairstyle simulation method according to the fourth aspect of the present invention can be executed by a computer capable of deforming only a part of a model image showing a hairstyle that is designated by a user. In a possible hairstyle simulation method, the computer includes a step of capturing an image of a user's own face, a step of extracting a contour line of a face portion from the image data of the captured face, and a contour line of the extracted face portion. And a process of matching the contour line of the face portion extracted from the model image data indicating the hairstyle with the contour line, and matching the contour line to the face portion of the image data of the captured face and the hair of the model image data. And a composite image is generated by superimposing the same on the touch display, and the hair part of the model image data is a polygon with texture represented by a node-link structure by a quadrangle primitive. A step of detecting a predetermined operation on a part of the hair part of the composite image which is displayed and displayed on the touch display, and a shape of the hair part is determined according to the detected predetermined operation. a step of deforming seen including, in the step of detecting the predetermined operation, change area as the predetermined operation, detects a moving direction and a moving amount, in the step of deforming the shape of the portion of the hair was detected For the changed area, the deformation direction and deformation amount of the textured polygon represented by the node-link structure of the quadrangle primitive are calculated based on the detected movement direction and movement amount, and the polygon deformation method and deformation amount are calculated. In the calculation step, auxiliary points are provided inside the quadrilateral primitive, and textured polygons represented by a node-link structure of four triangular primitives having the auxiliary points as vertices are gradually deformed while interlocking with each other. Characterize.

In the hairstyle simulation method according to a fifth aspect of the present invention, in the fourth aspect of the present invention, it is preferable that the computer detect a predetermined range from a coordinate position of the composite image displayed on the touch display as a change area.

Also, in the hairstyle simulation method according to the sixth invention, in the fourth or fifth invention , the computer sets a maximum deformation amount of a node forming each primitive to a distance from a contour line of a face portion of each node. Accordingly, it is preferable to set such that the shorter the distance, the smaller the distance.

Next, in order to achieve the above object, the computer program according to the seventh invention can be executed by a computer capable of transforming only a part of a model image showing a hairstyle which is designated by a user. In the computer program, the computer controls the image capturing means for capturing the face of the user himself, the contour extracting means for extracting the contour line of the face portion from the image data of the captured face, and the contour line of the extracted face portion for the hair. Matching means for performing matching processing with the contour line of the face part extracted from the model image data indicating the style, and the face part of the image data of the captured face and the hair part of the model image data by matching the contour line And the composite image are generated by superimposing them on the touch display to function as a composite image display means, and the hair portion of the model image data is a polygon with a texture represented by a node / link structure by a quadrangle primitive. Is displayed as, the computer, in accordance with the operation detection means for detecting a predetermined operation on a part of the hair portion of the composite image displayed on the touch display, and the detected predetermined operation, The operation detecting means is caused to function as an image deforming means for deforming the shape of the hair portion, the operation detecting means is caused to function as a means for detecting a change region, a moving direction and a moving amount as the predetermined operation, and the image deforming means is detected. With respect to the changed area, based on the detected movement direction and movement amount, it is caused to function as deformation calculation means for calculating the deformation direction and the deformation amount of the polygon with texture represented by the node-link structure of the quadrangle primitive, The deformation calculating means is a means for providing auxiliary points inside a quadrilateral primitive, and gradually deforming textured polygons represented by a node-link structure of four triangular primitives having the auxiliary points as vertices while interlocking with each other. and wherein the Rukoto to function.

Further, the computer program according to the eighth invention, in the seventh invention , causes the operation detecting means to function as a means for detecting a predetermined range from a coordinate position of the composite image displayed on the touch display as a change area. Is preferred.

Further, a computer program according to a ninth invention is the computer program according to the seventh or eighth invention , wherein the deformation calculating means sets a maximum deformation amount of a node constituting each primitive to a distance from a contour line of a face portion of each node. Accordingly, it is preferable to function as a means for setting such that it becomes smaller as the distance becomes shorter.

In the first invention, the fourth invention, and the seventh invention, the face of the user himself is imaged, the contour line of the face part is extracted from the image data of the imaged face, and the contour line of the extracted face part is extracted. Matching processing is performed with the contour line of the face portion extracted from the model image data indicating the hairstyle. By matching the contour lines, the face part of the imaged face image data and the hair part of the model image data are overlapped to generate a composite image, which is displayed on the touch display. The hair part of the model image data is displayed as a textured polygon expressed by a node-link structure with a quadrangle primitive, and a predetermined operation is performed on a part of the hair part of the composite image displayed on the touch display. Is detected, and the shape of the hair portion is deformed in accordance with the detected predetermined operation. As a result, only a part of the hair of the model image data can be transformed into a shape that does not have any unnaturalness. The hairstyle can be adjusted to the desired shape of the user.
In addition, the change area, the moving direction, and the movement amount are detected as a predetermined operation, and the detected change area is provided with a texture represented by a node / link structure using a quadrangle primitive based on the detected moving direction and the movement amount. Since the deformation direction and the amount of deformation of the polygon are calculated, it is possible to detect in which direction and to what extent the hair part of the composite image is deformed, and to prevent large deformation even in an unnecessary area. It becomes possible.
Further, since auxiliary points are provided inside the quadrilateral primitive and the textured polygons represented by the node-link structure of four triangular primitives with the auxiliary points as vertices are gradually deformed while interlocking with each other, so-called triangular primitives are used. As in the case of deforming a polygon with a texture represented by a node / link structure, there is no bend in the texture after deformation, and there is no unnatural part in the texture display. It is possible to deform it.

In the second invention, the fifth invention, and the eighth invention, a predetermined range is detected as a change area from the coordinate position of the composite image displayed on the touch display. As a result, it is possible to easily grasp the changed area of the hair portion of the composite image on the touch display, and it is possible to prevent a large deformation even in an area where the deformation is unnecessary.

In the third invention, the sixth invention, and the ninth invention, the maximum deformation amount of the node forming each primitive is set to be smaller as the distance is shorter, according to the distance from the contour line of the face part of each node. To do. As a result, even when the hair part is deformed in the direction of the face part, the hair part is not deformed beyond the contour line of the face part, and the hairstyle is deformed into a natural shape. It becomes possible.

  According to the present invention, only a part of the hair part of the model image data can be transformed into a shape without unnaturalness, so that the part that does not need to be adjusted is kept in the current state, Only, it is possible to adjust the hairstyle to the desired shape of the user.

It is a schematic diagram which shows the structure of the hairstyle simulation apparatus which concerns on embodiment of this invention. It is a functional block diagram which shows the structure of the hairstyle simulation apparatus which concerns on embodiment of this invention. It is a flow chart which shows the procedure of outline matching processing of CPU of a hairstyle simulation device concerning an embodiment of the invention. It is an image figure of superimposing the image of the face part of the hairstyle simulation device which concerns on embodiment of this invention. It is an image figure of the expansion / contraction of the image of the face part of the hairstyle simulation device which concerns on embodiment of this invention. It is an illustration figure of polygon deformation of a hair part. It is a comparison image figure before and after performing a rendering process with respect to a rectangular primitive. It is a comparison image figure before and after performing a rendering process with respect to a rectangular primitive. It is an example figure of the primitive used as the target of a rendering process. It is a comparison diagram of the result of the rendering process. It is a figure which shows the state after performing the rendering process with respect to two triangular primitives. It is a figure which shows the state after performing the rendering process with respect to four triangular primitives which made the auxiliary point P a vertex. It is a flowchart which shows the procedure of the hair image deformation process of CPU of the hairstyle simulation apparatus which concerns on embodiment of this invention.

  Hereinafter, a hairstyle simulation device according to an embodiment of the present invention will be described with reference to the drawings. Throughout the drawings to be referred to, elements having the same or similar configurations or functions are designated by the same or similar reference numerals, and detailed description thereof will be omitted.

  FIG. 1 is a schematic diagram showing the configuration of a hairstyle simulation device according to an embodiment of the present invention. As shown in FIG. 1, the hairstyle simulation device 1 according to the present embodiment is a terminal device such as a tablet having a touch display 16.

  The hairstyle simulation device 1 according to the embodiment of the present invention includes at least a CPU (central processing unit) 11, a memory 12, a storage device 13, an I / O interface 14, a video interface 15, a touch display 16, an imaging unit 17, and communication. It is composed of an interface 18 and an internal bus 19 that connects the above hardware.

  The CPU 11 is connected to each of the above-described hardware parts of the hairstyle simulation device 1 via the internal bus 19, controls the operation of each of the above-mentioned hardware parts, and stores the computer program 100 stored in the storage device 13. To perform various software functions. The memory 12 is composed of a volatile memory such as SRAM or SDRAM, and a load module is expanded when the computer program 100 is executed, and temporary data or the like generated when the computer program 100 is executed is stored.

  The storage device 13 includes a built-in fixed storage device (hard disk), ROM, and the like. The computer program 100 stored in the storage device 13 is downloaded from an external computer connected via the communication interface 18, and is expanded from the storage device 13 to the memory 12 and executed at the time of execution. Of course, the computer program may be read from the portable recording medium when the reader / writer is provided from the portable recording medium such as an SD card in which information such as programs and data is recorded.

  The storage device 13 stores model image data indicating a hairstyle. The model image data includes the contour data of the standard face portion, and the matching processing is executed with the contour data of the face portion extracted from the image data of the captured user's face. As a result, it is possible to generate a composite image of a face in which the hairstyle is deformed (adjusted) into a shape desired by the user.

  The communication interface 18 is connected to the internal bus 19, and by connecting to an external network, it is possible to perform data transmission / reception with an external computer or the like.

  The I / O interface 14 and the video interface 15 are connected to the touch display 16 composed of a liquid crystal display, and the face portion of the model image data showing the hairstyle and the face portion of the image data of the captured user's face. By executing the matching processing of the contour line data between and, the composite image generated by superimposing the face portion of the image data of the user's face and the hair portion of the model image data is displayed. By performing a swipe operation or the like on the displayed combined image, a part of the hair portion of the combined image is deformed.

  The image pickup means 17 is a CCD camera or the like and may or may not be provided. When the image pickup unit 17 is provided, a user's face is imaged, and a hair portion of the model image data indicating the hairstyle is combined with the face portion of the image data of the captured face to obtain a desired hairstyle. You can check the deformed state.

  The operation of the hairstyle simulation device 1 having the above configuration will be described below. FIG. 2 is a functional block diagram showing the configuration of the hairstyle simulation device 1 according to the embodiment of the present invention.

  In FIG. 2, the imaging unit (imaging unit) 201 of the hairstyle simulation device 1 according to the present embodiment images the face of the user who simulates the hairstyle. The image data of the face captured by the image capturing unit 201 is stored in the storage device 13. A CCD image sensor, a CMOS image sensor, or the like is used as the image pickup unit 201.

  The model image reading unit 202 reads the model image data stored in the storage device 13. The model image data is image data indicating a case where a standard model is set to the hairstyle ID in association with the hairstyle ID for identifying the hairstyle.

  The contour extracting unit 203 extracts the contour line of the face portion from the image data of the face captured by the image capturing unit 201. The method of extracting the contour line is not particularly limited. For example, a method of extracting a contour line when the pixel value of a predetermined pixel is separated from the pixel values of surrounding pixels by a predetermined value or more may be used. Similarly, with respect to the read model image data, the contour line of the face portion is extracted. As for the model image data, the contour line of the face portion may be extracted in advance, and the information regarding the extracted contour line may be stored together with the model image data.

  The matching unit 204 performs matching processing on the contour line of the extracted face portion with the contour line of the face portion extracted from the model image data indicating the hairstyle. FIG. 3 is a flowchart showing a procedure of contour matching processing of the CPU 11 of the hairstyle simulation device 1 according to the embodiment of the present invention.

  In FIG. 3, the CPU 11 of the hairstyle simulation device 1 according to the present embodiment calculates the center coordinates of the face portion from the imaged face image data and model image data (step S301). Specifically, a continuous area having a pixel value within a certain range is detected, and the center position is calculated as the vertical and horizontal centers of the continuous area.

  The CPU 11 moves the face part of the image data of the captured face so that the center coordinates of the face part of the image data of the captured face are superimposed on the center coordinates of the face part of the model image data ( Step S302). FIG. 4 is an image diagram of superimposing images of a face portion of the hairstyle simulation device 1 according to the embodiment of the present invention. Image pickup is performed so that the center coordinates G1 of the face portion of the imaged face image data shown in FIG. 4A is superimposed on the center coordinates G0 of the face portion of the model image data shown in FIG. 4B. The face portion of the image data of the face thus formed is moved.

  Returning to FIG. 3, the CPU 11 enlarges / reduces the face portion of the image data of the captured face in accordance with the height of the face portion of the model image data (step S303). For example, enlargement / reduction is performed so that the distance between the eyebrow and the jaw of the face portion of the model image data and the distance between the eyebrow and the jaw of the face portion of the imaged face image data match. At this time, the center of enlargement / reduction is the center coordinate of the calculated face portion. Also, the aspect ratio should be constant when enlarging / reducing.

  FIG. 5 is an image diagram of enlargement / reduction of the image of the face portion of the hairstyle simulation device 1 according to the embodiment of the present invention. As shown in FIG. 5, the face portion (broken line portion) of the imaged face image data is aligned with the face portion (solid line portion) of the model image data by superimposing the center coordinates.

  In the example of FIG. 5, the distance H1 between the eyebrow and the jaw of the face portion (broken line portion) of the imaged face image data is closer to the eyebrow and jaw of the face portion (solid line portion) of the model image data. Since the distance is longer than the distance H0 between H and H, the face portion of the image data of the imaged face is reduced while maintaining the aspect ratio until the distance H1 becomes the distance H0. This completes the matching process in the height direction.

  Returning to FIG. 3, the CPU 11 horizontally expands / reduces the face part of the model image data according to the width of the face part of the image data of the imaged face that has been enlarged / reduced (step S304). The horizontal displacement is smaller than the vertical displacement, and since the distance between the eyebrows and the jaws is matched by enlarging / reducing in the vertical direction, it is not necessary to strictly maintain the aspect ratio. Rather, the hairstyle of the model image data can be naturally combined with the face portion of the image data of the imaged face.

  The CPU 11 calculates the contour lines of the face portion of the imaged face image data and the face portion of the model image data by spline interpolation or the like (step S305), and calculates the lateral difference between the face contour lines ( Step S306). The CPU 11 moves the node of the hair part of the model image data by the calculated lateral difference (step S307). As a result, even if the contours of the face are different, it is possible to perform a simulation on a hairstyle shape that is not unnatural.

  Returning to FIG. 2, the composite image display unit 205 generates a composite image of the face portion of the captured image data of the face after the matching processing and the hair portion of the model image data, and displays the composite image on the touch display. .. The image of the hair portion is displayed as a polygon with a texture represented by a node-link structure using a quadrilateral primitive that uses a texture for displaying the hair. As a result, only the hair portion can be deformed and displayed by the following operation.

  The operation detection unit 206 detects a predetermined operation accepted by the user. Specifically, by detecting a swipe operation or the like on the touch display 16, the polygon with texture represented by the node / link structure of the quadrilateral primitive within a certain range is deformed. Thereby, the shape of the hairstyle can be finely adjusted by deforming only a part of the hair.

  As the predetermined operation to be detected, the changed area, the moving direction and the moving amount by the swipe operation are detected. As the change area, an area within a certain range from the center of the pixel group in which the contact of the finger is detected is detected as the change area. The changed area may be enlarged / reduced stepwise according to the number of pixel groups. The moving direction is calculated based on the moving direction of the finger while swiping, and the moving amount is calculated based on the moving amount that the finger moves while touching while swiping.

  The image transformation unit 207 transforms the shape of the hair portion of the composite image according to the detected predetermined operation. Further, with respect to the detected changed area, the polygon deformation is calculated by performing the rendering process on the quadrangle primitive based on the detected moving direction and moving amount, and the accompanying texture is deformed.

  FIG. 6 is a view showing an example of polygon deformation of a hair part. As shown in FIG. 6, grid-shaped nodes 61 that are linked to each other are formed, and a composite image is pasted thereon as a texture. By the detected operation, each of the grid-like nodes is deformed, and the linked nodes are sequentially deformed accordingly, so that the entire changed region is deformed. For example, as shown in the change areas 62 and 63, the node 61 is deformed, and the texture of the attached hair portion is deformed, so that the hair style can be freely changed.

  That is, the image transforming unit 207 calculates the transforming direction and the transforming amount of the textured polygon represented by the node-link structure by the quadrangle primitive, based on the detected moving direction and the moving amount for the detected changed region. The transformation calculation unit 208 is provided. Accordingly, by obtaining the deformation direction and the deformation amount of the polygon, it is possible to deform the image of the hair part that is attached as a texture. The composite image display unit 205 updates the composite image based on the calculated deformation direction and deformation amount, and displays it on the touch display 16.

  In addition, in this embodiment, since the nodes are formed in a grid pattern, the rendering process is executed on the quadrilateral primitive. FIG. 7 is a comparison image diagram before and after performing the rendering process on the quadrangle primitive. FIG. 7A shows the state of the quadrilateral primitive before the execution of the rendering process, and FIG. 7B shows the state of the quadrilateral primitive after the execution of the rendering process.

  In the example of FIG. 7A, three quadrilateral primitives are arranged in series. Straight lines T0-T2 are drawn as textures on the three quadrilateral primitives. Then, the rendering process is executed so that the node N1 becomes the node N2.

  In this case, the straight line is transformed into a straight line inside the square primitive. Therefore, as shown in FIG. 7B, when the node N1 is pulled to the position of the node N2, the texture position T1 moves to T1 'and the texture position T2 moves to T2'. As a result, texture mapping is performed as a relatively angular texture.

  On the other hand, in a smart phone or the like, hardware is restricted, so that OpenGL cannot be used and OpenGL ES 2.0 may be used. In this case, the rendering process cannot be executed on the rectangular primitive.

  Therefore, the rendering process may be executed by dividing the quadrilateral primitive into a plurality of triangle primitives. In this case, a method of dividing into two with one diagonal line and a method of dividing into four with two diagonal lines can be considered. FIG. 8 is an exemplary diagram of primitives that are the target of the rendering process.

  In FIG. 8A, the quadrilateral primitive is divided into two by one diagonal to form two triangle primitives. In FIG. 8B, the inside of the quadrilateral primitive is divided into four triangular primitives having the auxiliary point P at the intersection of two diagonal lines and the auxiliary point P as a vertex.

  Although either method can perform the rendering process on the triangle primitive, the result is different. FIG. 9 is a comparison diagram of the results of the rendering process. FIG. 9A schematically shows the result of rendering processing performed on the triangle primitive of FIG. 8A, and FIG. 9B schematically shows the result of rendering processing performed on the triangle primitive of FIG. 8B. FIG.

  As can be seen by comparing FIG. 9A and FIG. 9B, in FIG. 9A, a clear crease is generated in the texture, and the texture after deformation as a whole is uncomfortable. On the other hand, in FIG. 9B, creases are unlikely to occur in the texture, and the texture after deformation is unlikely to feel unnatural as a whole. Therefore, it can be seen that the deformed hairstyle has a more natural shape.

  A simple example explains the difference in the texture mapping results. FIG. 10 is a diagram (corresponding to FIG. 8A) showing a state after the rendering process is performed on the two triangular primitives. 10A and FIG. 10B, the difference after the rendering process is executed for the triangle primitive divided by each of the two diagonal lines of the quadrangle primitive can be seen.

  That is, in FIG. 10A, the quadrilateral primitive is divided into two triangular primitives by diagonal lines from the upper right to the lower left, and the direction of the diagonal line is the moving direction from the node N1 to the node N2 (see FIG. 7). The directions are intersecting. In this case, the bending of the texture after the affine transformation due to the triangular deformation is relatively large (see texture positions T3 and T4), and sharp folds are generated in the texture.

  On the other hand, in FIG. 10B, the quadrangle primitive is divided into two triangular primitives by diagonal lines from the upper left to the lower right, and the direction of the diagonal line is the moving direction from the node N1 to the node N2 (see FIG. 7). The directions are almost the same. In this case, the amount of bending of the texture after affine transformation due to the triangular deformation is relatively small (see texture position T5), and creases are unlikely to occur in the texture.

  In the case of dividing into two triangle primitives in this way, a result may be that a large crease may or may not occur in the texture depending on the relative relationship between the diagonal direction and the deforming direction. Therefore, there is a possibility that an unnatural texture will occur.

  On the other hand, FIG. 11 is a diagram (corresponding to FIG. 8B) showing a state after the rendering process is executed for the four triangular primitives having the auxiliary point P as a vertex. In FIG. 11, the intersection of two diagonal lines is added as a new node (auxiliary point P) and divided into four triangular primitives.

  Then, the rendering process is executed on the four triangular primitives. In this case, since the method of dividing the quadrilateral primitive is symmetrical, the bending of the texture after affine transformation due to the triangular deformation converges within a certain range without being influenced by the moving node or the moving direction of the node. (See texture positions T6 and T7). Therefore, sharp folds are unlikely to occur in the texture, and natural deformation can be performed.

  It is preferable to set the maximum deformation amount of the nodes forming the quadrangle primitive or the triangle primitive such that it becomes smaller as the distance is shorter, according to the distance from the contour line of the face portion of each node. In this case, the deformation amount is 0 (zero) for the node on the contour line of the face portion. As a result, even when the hair part is deformed in the direction of the face part, the hair part is not deformed beyond the contour line of the face part, and the hairstyle is deformed into a natural shape. It becomes possible.

  FIG. 12 is a flowchart showing the procedure of the hair image transformation process of the CPU 11 of the hairstyle simulation device 1 according to the embodiment of the present invention. The CPU 11 of the hairstyle simulation device 1 according to the present embodiment images the face of the user who simulates the hairstyle (step S1201). The image data of the captured face is stored in the storage device 13.

  The CPU 11 reads out the model image data stored in the storage device 13 (step S1202). The model image data is image data indicating a case where a standard model is set to the hairstyle ID in association with the hairstyle ID for identifying the hairstyle.

  The CPU 11 extracts a contour line of the face from the imaged face image data (step S1203), and also extracts a contour line of the face portion from the read model image data (step S1204). As for the model image data, the contour line of the face portion may be extracted in advance, and the information regarding the extracted contour line may be stored together with the model image data.

  The CPU 11 performs matching processing on the extracted contour line of the face portion with the contour line of the face portion of the model image data indicating the hairstyle (step S1205), and the face of the image data of the imaged face after the matching processing is performed. A composite image of the portion of (1) and the hair portion of the model image data is generated and displayed on the touch display 16 (step S1206). The image of the hair portion is displayed as a polygon with a texture represented by a node-link structure using a quadrilateral primitive that uses a texture for displaying the hair.

  The CPU 11 detects a predetermined operation accepted by the user (step S1207). Specifically, by detecting a swipe operation or the like on the touch display 16, the polygon with texture represented by the node / link structure of the quadrilateral primitive within a certain range is deformed.

  As the predetermined operation to be detected, the changed area, the moving direction and the moving amount by the swipe operation are detected. As the change area, an area within a certain range from the center of the pixel group in which the contact of the finger is detected is detected as the change area. The changed area may be enlarged / reduced stepwise according to the number of pixel groups. The moving direction is detected as the moving direction of the finger while swiping, and the moving amount is detected as the moving amount of the finger while swiping.

  The CPU 11 calculates the deformation direction and the deformation amount of the textured polygon represented by the node-link structure of the quadrangle primitive, based on the detected movement direction and movement amount of the detected change area (step S1208). .. Specifically, a polygon deformation is calculated by performing a rendering process on a quadrilateral primitive, and the associated texture is deformed. The CPU 11 updates the composite image based on the calculated deformation direction and the calculated deformation amount and displays it on the touch display 16 (step S1209).

  As described above, according to the present embodiment, only a part of the hair part of the model image data can be deformed into a shape without unnaturalness, so that the current condition is maintained for the part that does not need to be adjusted. As it is, it is possible to simulate only the part of interest, and it becomes possible to adjust the hairstyle to the shape desired by the user.

  It should be noted that the present invention is not limited to the above-described embodiments, and various modifications and improvements can be made within the scope of the gist of the present invention. For example, the method of detecting the changed area is not particularly limited, and, for example, contact with a plurality of fingers may be detected, and an area including both contact areas may be calculated as the changed area.

1 Hairstyle simulation device 11 CPU
12 memory 13 storage device 17 imaging means 100 computer program P auxiliary point

Claims (9)

In a hairstyle simulation device capable of deforming only a part of the model image showing the hairstyle that has been specified by the user,
Imaging means for imaging the user's own face,
A contour extracting means for extracting a contour line of a face portion from the imaged face image data;
Matching means for matching the contour line of the extracted face portion with the contour line of the face portion extracted from the model image data indicating the hairstyle,
By combining the contour lines, the face part of the captured image data of the face and the hair part of the model image data are overlapped to generate a composite image, and the composite image display means is displayed on the touch display. Have,
The hair part of the model image data is displayed as a polygon with a texture represented by a node / link structure using a quadrangle primitive,
Operation detecting means for detecting a predetermined operation on a part of the hair portion of the composite image displayed on the touch display;
Image deforming means for deforming the shape of the hair portion according to a predetermined operation detected ,
The operation detecting means detects a change area, a moving direction and a moving amount as the predetermined operation,
The image transforming means,
Deformation calculation means for calculating the deformation direction and the deformation amount of the textured polygon represented by the node-link structure of the quadrilateral primitive, based on the detected movement direction and movement amount of the detected change area.
Equipped with
The deformation calculating means provides auxiliary points inside a quadrilateral primitive, and gradually deforms textured polygons represented by a node-link structure of four triangular primitives having the auxiliary points as vertices while interlocking with each other. Hairstyle simulation device featuring. The hairstyle simulation device according to claim 1, wherein the operation detection unit detects a predetermined range from a coordinate position of the composite image displayed on the touch display as a change region . The deformation calculation means sets the maximum deformation amount of a node forming each primitive such that the smaller the distance is, the smaller the maximum deformation amount of the node is, depending on the distance from the contour line of the face portion of each node. Item 3. The hairstyle simulation device according to item 1 or 2. In a hairstyle simulation method that can be executed by a computer that can deform only a part of a model image showing a hairstyle that has been specified by the user,
The computer is
Capturing the user's own face,
A step of extracting a contour line of a face portion from the imaged face image data;
A process of matching the contour line of the extracted face portion with the contour line of the face portion extracted from the model image data indicating the hairstyle,
A step of superimposing the face part of the image data of the imaged face and the hair part of the model image data by matching the contour lines to generate a composite image, and displaying the composite image on the touch display.
Including,
The hair part of the model image data is displayed as a polygon with a texture represented by a node / link structure using a quadrangle primitive,
Detecting a predetermined operation on a part of the hair portion of the composite image displayed on the touch display;
A step of deforming the shape of the hair part according to a predetermined operation detected;
Including,
In the step of detecting the predetermined operation, the change area, the moving direction and the moving amount are detected as the predetermined operation,
In the step of deforming the shape of the hair part, with respect to the detected change area, the deformation direction of the textured polygon expressed by the node / link structure by the quadrangle primitive is detected based on the detected movement direction and movement amount. Calculate the amount of deformation,
In the step of calculating the deformation method and the deformation amount of the polygon, an auxiliary point is provided inside the quadrilateral primitive, and a polygon with a texture represented by a node-link structure by four triangular primitives having the auxiliary point as a vertex, It features and to Ruhe a style simulation method to be gradually deformed while interlocked with each other. The hairstyle simulation method according to claim 4, wherein the computer detects a predetermined range from the coordinate position of the composite image displayed on the touch display as a change area . The computer according to claim 4 in which the maximum deformation amount of nodes constituting each primitive, in accordance with the distance from the contour of the portion of the face of each node, distances and sets so that shorter smaller Alternatively, the hairstyle simulation method according to item 5 . In a computer program that can be executed by a computer capable of transforming only a part of a model image showing a hairstyle that has been designated by the user,
The computer ,
An image capturing means for capturing the user's own face,
Contour extraction means for extracting the contour line of the face portion from the imaged face image data,
Matching means for performing a matching process on the contour line of the extracted face portion with the contour line of the face portion extracted from the model image data indicating the hairstyle;
Synthetic image display means for generating a synthetic image by superimposing the face portion of the imaged face image data and the hair portion of the model image data by matching the contour lines, and displaying the synthesized image on the touch display.
Function as
The hair part of the model image data is displayed as a polygon with a texture represented by a node / link structure using a quadrangle primitive,
The computer,
Operation detecting means for detecting a predetermined operation on a part of the hair portion of the composite image displayed on the touch display, and
Image deforming means for deforming the shape of the hair portion in accordance with the detected predetermined operation
Function as
Causing the operation detecting means to function as a means for detecting a change area, a moving direction and a moving amount as the predetermined operation,
Deformation for calculating the deformation direction and the deformation amount of the textured polygon represented by the node-link structure by the quadrangle primitive based on the detected movement direction and movement amount of the detected change area with respect to the detected change region. Make it function as a calculation means,
The deformation calculating means is for providing auxiliary points inside a quadrilateral primitive, and gradually deforming textured polygons represented by a node-link structure of four triangular primitives having the auxiliary points as vertices while interlocking with each other. computer program that is characterized in that to function as a. 8. The computer program according to claim 7, wherein the operation detecting means is caused to function as a means for detecting a predetermined range from a coordinate position of the composite image displayed on the touch display as a change area . Causing the deformation calculation means to function as means for setting the maximum deformation amount of the node forming each primitive so that it becomes smaller as the distance becomes shorter, according to the distance from the contour line of the face portion of each node. Computer program according to claim 7 or 8, characterized in that