JP2011039828A - System and method for processing image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing system allowing higher-level two-dimensional display by use of a three-dimensional model. <P>SOLUTION: When a viewpoint position VP4 to a character A is present in an interpolation viewpoint direction VDi2, two model viewpoint directions VDm1, VDm2 constituting an interpolation range INr including a viewpoint direction VDi are specified. By interpolating each vertex forming a front model M1 corresponding to one model viewpoint direction VDm1 and each vertex forming a 45° model M2 corresponding to the other model viewpoint direction VDm2, an interpolation image Ima-i2 when viewing the character A from the viewpoint position VP4 is displayed on a monitor 3. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing system and an image processing method for displaying an image based on a three-dimensional model.

  An image processing system that generates one three-dimensional model during modeling and determines an image to be displayed on a monitor based on the three-dimensional model generated during rendering and the viewpoint direction is well known. However, when displaying an image of a character known in manga or animation, which is a two-dimensional medium, based on the three-dimensional model, it is expressed in two dimensions only by faithfully displaying the three-dimensional model viewed from the viewpoint position. In some cases, the display characteristics of the character that has been lost may be impaired. To solve this problem, a method is proposed in which a desired reference image viewed from the reference direction is generated in advance, and the character object arranged in the reference image is shifted according to the deviation between the viewpoint direction and the reference direction. (For example, refer to Patent Document 1).

JP 2007-1567773 A

  However, the above processing method only shifts the shape of the object provided in the reference image with respect to the reference direction. For example, when the shape changes completely with the change of the viewpoint direction, I can't express changes.

  Therefore, an object of the present invention is to provide an image processing system and an image processing method that enable a more advanced two-dimensional display using a three-dimensional model.

  The present invention solves the above-described problems by the following means. In order to facilitate understanding of the present invention, reference numerals in the accompanying drawings are appended in parentheses, but the present invention is not limited to the illustrated embodiment.

  The first image processing system (1) according to the present invention is based on a three-dimensional model (M) of a display target object (A) arranged in a virtual three-dimensional space (GS). The image processing system (1) displays an image of the display object on the display unit (3) according to a viewpoint direction (VD) that is a direction of VP), and the viewpoint position is the first with respect to the display object. First model data (MD1) for forming a first model (M1) to be used as a three-dimensional model of the display object when it is in one viewing direction (VDm1), and the display object As a three-dimensional model of the display object when the viewpoint position is in the second viewpoint direction (VDm2) in which the first viewpoint direction is moved in a predetermined circumferential direction around the display object. Form the second model (M2) used Model data storage means (4) for storing the second model data (MD2) for interpolation, and an interpolation viewpoint direction (VDi) existing between the first viewpoint direction and the second viewpoint direction with respect to the circumferential direction ), An interpolated image (INima), which is a two-dimensional image of the display object, is obtained from each vertex coordinate of the first model obtained from the first model data and from the second model data. Interpolated image generation means (5a) for generating an interpolation image based on each vertex coordinate of the second model and drawing the interpolation image in a predetermined drawing area, and the drawing drawn in the drawing area Interpolated image display means (5b) for displaying an interpolated image on the display unit as the display object viewed from the viewpoint position when the viewpoint position is in the interpolation viewpoint direction with respect to the display object. With the door, by, to solve the foregoing problems.

  According to the first image processing system of the present invention, the first model is set and the second viewpoint direction is set so as to obtain an aspect to be expressed when the display object is viewed from the first viewpoint direction. If the second model is set so that the aspect to be expressed when viewed from the viewpoint is obtained, the display object is expressed in the aspect to be expressed with respect to the first and second viewpoint directions. Can do. On the other hand, when the viewpoint position with respect to the display object is in the interpolation viewpoint direction between the first viewpoint direction and the second viewpoint direction, the first model data and the second model data for forming the first model By performing the interpolation process based on the second model data for forming the model, it is possible to generate an interpolation image when the display object is viewed from the interpolation viewpoint direction. Accordingly, it is possible to provide a two-dimensional expression using a three-dimensional model, and to naturally represent a mode change, and to provide an image processing system that enables a more advanced two-dimensional expression. be able to.

  The first model data and the second model data are a set of vertex coordinates for forming the corresponding model, and a function for obtaining the vertex coordinates of the corresponding model from the reference shape. including. The display object includes, for example, a case where there is a face like a character appearing in a game or animation, and a case where the display target has an inorganic three-dimensional shape without a face. In the interpolated image generating means, a two-dimensional image is obtained from each of the first model data and the second model data, and the coordinates in each two-dimensional image are interpolated to obtain the coordinates of the interpolated image. And interpolating the vertex coordinates of the model data and the second model data to generate a three-dimensional model and obtaining a two-dimensional image from the generated three-dimensional model.

  Each of the first and second model data includes association information (30) that associates each vertex forming the first model with each vertex forming the second model, The interpolation image generation means may generate the interpolation image by performing an interpolation process while maintaining the correspondence between the vertices. This facilitates interpolation processing.

  In each of the first and second model data, group identification information for identifying a corresponding group is associated with each vertex of the plurality of polygons so that each vertex forms a group, A group interpolation information storage unit (4) that stores an interpolation method used in the interpolation image generation unit in association with each group is further provided, and the interpolation image generation unit is associated with each group. The interpolation image may be generated by performing the interpolation processing based on each vertex coordinate corresponding to the group by the interpolation method. Thus, for example, when the display target is a character, if the parts of the character's hair, nose, mouth, eyes, etc. are set as a group, each part is interpolated by an interpolation method according to the part, and changes depending on the part. Interpolated images having different aspects can be obtained.

  Relative positions of the interpolation viewpoint direction set in the interpolation range between the first viewpoint direction and the second viewpoint direction, and each of the first viewpoint direction and the second viewpoint direction According to the relationship, the first model data and the second model data specify the first and second influence levels indicating the magnitude of the influence on the interpolation in the interpolated image generating means. The interpolated image generating means further includes the first influence degree and the second influence degree specified by the influence degree specifying means based on the first influence degree and the second influence degree. The interpolation processing based on each vertex coordinate obtained from the model data and each vertex coordinate obtained from the second model data may be performed.

  Thereby, in the aspect which changes from the first model to the second model, the influence of the first model or the second model is set to be large, or the influence of either one is set to be large. be able to. As a method for specifying the first and second influence degrees, for example, the relative positional relationship of the interpolation viewpoint direction with respect to each of the first viewpoint direction and the second viewpoint direction is associated with the influence degree. The degree of influence is obtained from the method of specifying by referring to the degree of influence table and the relative positional relationship of the interpolation viewpoint direction with respect to each of the first and second viewpoint directions. There is a method of specifying a function to be specified and specifying from the function.

  Operation input means (2) for receiving an input operation of the user is further provided, the first model is displayed on the display unit, and the first model displayed on the display unit is displayed according to the user operation. Cutout means (5c) for cutting out the specific part, correction means (5d) for accepting a correction operation by the user for the cut out specific part, and the first part in which the specific part has been corrected by the correction means You may have a 2nd model setting means (5e) which sets a model as said 2nd model.

  In the first image processing system of the present invention, it is necessary to prepare a plurality of three-dimensional models. However, according to the above invention, if one three-dimensional model is generated, the generated three-dimensional model is included in the generated three-dimensional model. Based on this, other three-dimensional models can be easily generated.

  A reference axis (CA) passing through the display object and extending in a predetermined direction is set, and in the model data storage unit, the second viewpoint direction is centered on the reference axis and the first viewpoint direction is a predetermined circumference. And the interpolation image generating means is provided with a viewpoint direction specifying means for specifying the direction of the viewpoint position with respect to the reference axis as the viewpoint direction with respect to the display object, and the specified viewpoint direction May exist between the first viewpoint direction and the second viewpoint direction, the interpolation image may be generated with the viewpoint direction as the interpolation viewpoint direction. Thereby, the direction of the viewpoint position with respect to the display object can be specified as the direction of the viewpoint position with respect to the reference axis passing through the display object, and the specification of the viewpoint direction of an arbitrary viewpoint position in the virtual three-dimensional space is prepared. .

  The model data storage means is a model for forming a three-dimensional model of the display object corresponding to each model viewpoint direction with respect to a plurality of model viewpoint directions set on a circumferential direction around the reference axis. Data (MD) is stored, and the interpolation image generation means specifies two model viewpoint directions sandwiched with respect to the circumferential direction when the viewpoint direction specified by the viewpoint direction specification means is the interpolation viewpoint direction. And further comprising model viewpoint direction setting means for setting one model viewpoint direction as the first viewpoint direction and the other model viewpoint direction as the second viewpoint direction, wherein the interpolated image generating means includes the model Based on the first viewpoint direction and the second viewpoint direction set by the viewpoint direction setting means, the interpolated image corresponding to the specified viewpoint direction is generated. It may be. Thereby, an interpolation image corresponding to an arbitrary interpolation viewpoint direction in the virtual three-dimensional space can be generated.

  Interpolation information storage means (4) for storing range information (RI) in which information relating to interpolation of two corresponding models is set for each interpolation range constituted by two model viewpoint directions adjacent to each other in the circumferential direction The model viewpoint direction specifying means specifies the two model viewpoint directions by specifying the range information of the interpolation range including the specified viewpoint direction with reference to the interpolation information storage means. Then, the interpolated image generating means may interpolate model data corresponding to the two models based on the information related to the interpolation. Thereby, a different interpolation method can be set according to the interpolation range, and a more appropriate mode of change can be expressed according to the angle at which the display object is viewed.

  The interpolation image generation means includes an interpolation model generation means for generating a three-dimensional interpolation model by interpolating each vertex coordinate obtained from the first model data and each vertex coordinate obtained from the second model data. Further, the previous three-dimensional interpolation model is arranged at the predetermined position as a three-dimensional model of the display object, and an image obtained from the three-dimensional interpolation model when the predetermined gazing point is viewed from the viewpoint position, You may produce | generate as the said interpolation image of a display target object.

  Thereby, it is possible to obtain an interpolated image of the display object to be expressed when an arbitrary gazing point is viewed from an arbitrary viewpoint position. Note that the image obtained when viewing the gazing point from the viewpoint position is an image obtained by projecting the three-dimensional model onto a plane that includes the viewpoint position and is perpendicular to the gaze vector that connects the viewpoint position and the gazing point.

  The interpolated image generation means sets a point that has the same distance from the viewpoint position and the reference axis and is in the set first viewpoint direction as a first temporary viewpoint position, and sets the set first viewpoint When the first model corresponding to the viewing direction is arranged at the predetermined position as a three-dimensional model of the display object and the predetermined gazing point is viewed from the first temporary viewpoint position, the first model The first temporary image acquisition means for acquiring the first temporary image obtained from the first temporary image acquisition means, and a point in the second viewpoint direction that has the same distance from the viewpoint position and the reference axis and in the set second viewpoint direction. A second position corresponding to the set second viewpoint direction is arranged at the predetermined position as a three-dimensional model of the display object, and the gazing point is determined from the second temporary viewpoint position. The second temporary obtained from the second model Second provisional image obtaining means for obtaining an image, and interpolating each vertex coordinate of the first provisional image and each vertex coordinate of the second provisional image according to the specified viewpoint direction. The image obtained by this may be generated as an interpolated image of the display body when the gazing point is viewed from a viewpoint position in the specified viewpoint direction with respect to the display object.

  Thereby, an interpolation image can be generated by interpolating the coordinates of the two-dimensional image. Note that the image of the three-dimensional model obtained when the gazing point is viewed from the temporary viewpoint position includes the temporary viewpoint position and is an image obtained by projecting the three-dimensional model onto a plane perpendicular to the gaze vector connecting the temporary viewpoint position and the gazing point. It is.

  A plurality of regions (VA) divided into a latitude direction (VM) and a meridian direction (VL) are set in a spherical virtual region (G) centered on a predetermined center point (CP) in the display object. The first model data includes the first model data as the three-dimensional model of the display object when the viewpoint position is in any viewpoint direction of the first region (VAm1) including the first viewpoint direction. The second model data is data for forming one model, and the second model data is a viewpoint direction of any one of the second regions (VAm2) positioned in the circumferential direction around the center point. Is the data for forming the second model as a three-dimensional model of the display object, and the interpolated image generation means is configured to output the first region from the first region with respect to the circumferential direction. Up to the second area When there is a viewpoint position in the interpolation viewpoint direction included in the interpolation region existing in the interpolation area, each of the vertex coordinates obtained based on the first model data, an interpolation image corresponding to the interpolation viewpoint direction, and the second Each vertex coordinate obtained based on the model data may be generated by interpolation and drawn in a predetermined drawing area.

  Thereby, even when a display object having a characteristic aspect is displayed with respect to all directions of the display object, an interpolation image can be efficiently generated.

  According to a first image processing method of the present invention, a viewpoint position (with respect to the display target object) is determined based on a three-dimensional model (M) of the display target object (A) arranged in a virtual three-dimensional space (GS). An image processing method for causing the display unit (3) to display an image of the display object according to a viewpoint direction (VD) that is a direction of VP), wherein the viewpoint position is a first viewpoint with respect to the display object. A first model data for forming a first model used as a three-dimensional model of the display target object, and a viewpoint position with respect to the display target object. A second model for forming a second model used as a three-dimensional model of the display object when the first viewpoint direction is a second viewpoint direction moved in a predetermined circumferential direction at the center; Storing model data; and An interpolation image, which is a two-dimensional image of the display object corresponding to the interpolation viewpoint direction existing between the first viewpoint direction and the second viewpoint direction with respect to the direction, is obtained from the first model data. A step of performing interpolation processing based on each vertex coordinate of the first model data and each vertex coordinate of the second model obtained from the second model data, and drawing it in a predetermined drawing area And displaying the interpolation image drawn in the drawing area on the display unit as the display object corresponding to the interpolation viewpoint direction. The first image processing method of the present invention is realized as the first image processing system of the present invention.

  The second image processing system (100) of the present invention is based on a three-dimensional model (M) of the display object (A) arranged at a predetermined position in the virtual three-dimensional space, An image processing system for displaying an image of the display object on the display unit (3) according to a viewpoint direction (VD) that is a direction of VP), wherein a viewpoint position with respect to the display object is in a predetermined reference direction In this case, a model data storage unit (104) that stores reference model data for forming a reference model used as a three-dimensional model of the display object, and a viewpoint position with respect to the display object indicates the reference direction. When the moving object is in the moving viewpoint direction moved in the circumferential direction around the display object, a moving model used as a three-dimensional model of the display object corresponding to the moving viewpoint direction is configured. Model forming function (F) set such that each vertex of the polygon is obtained based on each vertex coordinate of the plurality of polygons constituting the reference model and the movement amount of the moving viewpoint direction from the reference direction. ) Is stored according to the moving viewpoint direction, and the moving viewpoint direction stored in the function storage section is the viewpoint position at which the viewing object is viewed, The model forming function corresponding to the specified moving viewpoint direction is selected, and the moving model is placed at the predetermined position by the selected model forming function based on the moving amount and each vertex coordinate of the reference model. A moving model generating means (105a) for generating, and a drawing means (105b) for drawing a moving model image obtained by viewing the generated moving model from the viewpoint position in a predetermined drawing area; A drawing display means (105c) for displaying the moving model image drawn in the drawing area as an image of the display object when the display object is viewed from the viewpoint position; This solves the above-mentioned problem.

  According to the second image processing system, as model data for forming a three-dimensional model used for displaying a display object on a display unit, a reference model used when a viewpoint direction is in a reference direction. Only the model data is stored in the model data storage unit, and the moving model used when viewing the display object from an arbitrary moving viewpoint direction is generated by the model formation function, and the generated moving model is viewed from the moving viewpoint position. The image of the displayed object is drawn in the drawing area. By setting a model forming function so that a desired display mode can be obtained for each moving viewpoint direction, a model image having a desired display mode can be obtained based on one three-dimensional model. Accordingly, it is possible to provide a two-dimensional expression using a three-dimensional model, and to naturally express a change in an aspect, and to provide an image processing system that enables a more advanced two-dimensional expression. be able to.

  The reference model data is composed of vertex coordinates of a plurality of polygons that form a corresponding model. Each vertex coordinate includes a case where a relative value is set and a case where an absolute value is set. The display object includes, for example, a case where there is a face like a character appearing in a game or animation, and a case where the display target has an inorganic three-dimensional shape without a face.

  In the function storage unit, for each of a plurality of function sections provided in the circumferential direction of the display object, the movement based on the movement amount in the movement viewpoint direction and each vertex coordinate of the reference model The model formation function set so as to obtain each vertex coordinate of the model is associated, and the movement model generation means is configured to generate the selected model based on each vertex coordinate of the reference model and the movement amount. The movement model may be generated by a function. Thereby, the same function can be associated with the moving viewpoint directions included in the same function section.

  In the reference model data, vertex coordinates of a plurality of polygons constituting the reference model are set to be grouped into a plurality of groups. In the function storage unit, for each of the groups, the group corresponding to the group is set. A model forming function for obtaining each vertex coordinate of the movement model from each vertex coordinate of the reference model may be set. Thereby, when the display mode of the display object in the circumferential direction varies depending on the parts of the display target, the vertex coordinates included in the parts that change in the same way are set in the same group, and the display mode of the group for each group If the functions set so that can be expressed are associated, different display modes can be expressed depending on the parts.

  A reference axis (CA) extending through the display object and extending in a predetermined direction is set, the circumferential direction is a direction centered on the reference axis, and the movement model generation means is a direction of a viewpoint position with respect to the reference axis A viewpoint direction specifying means (105d) for specifying as a viewpoint direction for the display object, specifying the viewpoint direction specified by the viewpoint direction specifying means as the moving viewpoint direction, and generating the movement model May be. Thereby, for example, the moving viewpoint direction of the viewpoint position arbitrarily determined by a predetermined application can be specified as the direction with respect to the reference axis.

  The second image processing method of the present invention is used as a three-dimensional model (M) of the display object when the viewpoint position (VP) with respect to the display object (A) is in a predetermined reference direction (VDc). A model data storage unit (104) for storing reference model data (MD ′) for forming a reference model (Mc), and a viewpoint position with respect to the display target object, the reference direction being centered on the display target object When there is a moving viewpoint direction (MVD) moved in the circumferential direction, each vertex of a plurality of polygons constituting a moving model used as a three-dimensional model of the display object corresponding to the moving viewpoint direction is A model forming function (F) set so as to be obtained based on each vertex coordinate of a plurality of polygons constituting the reference model and a movement amount of the moving viewpoint direction from the reference direction The display object based on a three-dimensional model of the display object arranged at a predetermined position in a virtual three-dimensional space (GS) in a computer having a function storage unit (104) for storing according to a point direction An image processing method for causing the display unit (3) to display an image of the display object according to a viewpoint direction that is a direction of a viewpoint position with respect to the image, wherein the viewpoint position at which the computer views the display object is the function Identify which moving viewpoint direction is stored in the storage unit, select the model forming function corresponding to the specified moving viewpoint direction, based on the movement amount and each vertex coordinate of the reference model, A step of generating the movement model at the predetermined position by the selected model forming function; and a movement model image obtained by viewing the generated movement model from the viewpoint position. Drawing in the drawing area; and causing the display unit to display the movement model image drawn in the drawing area as an image of the display object when the display object is viewed from the viewpoint position. By executing this, the above-mentioned problem is solved. The second image processing method of the present invention is embodied as a second image processing system by causing a computer to execute the second image processing method.

  As described above, according to the present invention, the first model data for forming the first model used when displaying the display object when the display object is viewed from the first viewpoint direction. And forming a second model used when expressing the display object when viewed from the second viewpoint direction in which the first viewpoint direction is moved in the circumferential direction around the display object. Using the three-dimensional model by preparing second model data and obtaining an interpolation image corresponding to the interpolation viewpoint direction existing between the first viewpoint direction and the second viewpoint direction by interpolation processing It is possible to provide an image processing system and an image processing method that enable more advanced two-dimensional display. Also, the display target when viewed from the moving viewpoint direction based on the reference model data for forming the reference model used to display the display target when the display target is viewed from the predetermined reference direction Generate a moving model used to display the body, and configure the moving model to obtain an image when viewed from the moving viewpoint direction. An image processing system and an image processing method can be provided.

The bird's-eye view of the character arrange | positioned in the image generation space in a 1st form. The figure which shows the front character image which looked at the character from the front. The figure which shows the 45 degree character image which looked at the character from 45 degree | times on the right. . The figure which shows the 90 degree character image which looked at the character from 90 degree | times on the left. The figure which shows the example of an interpolation viewpoint direction and an interpolation range. FIG. 6 is a diagram showing a state where a plurality of interpolation viewpoint directions are set in one interpolation range. The figure which shows the example of two influence degree tables. The figure which shows the procedure which produces | generates the two-dimensional image of a character regarding a model viewpoint direction. The figure which shows the procedure which produces | generates the interpolation image of a character based on the 1st image generation method in a 1st form. The figure which shows the procedure which produces | generates the interpolation image of a character based on the 2nd image generation method in a 1st form. The figure which shows an example of the hardware structure of the image processing system in a 1st form. The figure which shows the data structure of the model data in a 1st form. The figure which shows the data structure of interpolation information. The flowchart which shows the flow of the process in a modeling process. The flowchart which shows the flow of the process in the rendering process in a 1st form. The figure which shows so that a character image may be produced | generated based on the 1st image production | generation method in the rendering process in a 1st form. The figure which shows so that a character image may be produced | generated based on the 2nd image production | generation method in the rendering process in a 1st form. The figure which shows another form regarding the setting of the reference axis in a 1st form. The figure which shows the relationship between a virtual area and a direction vector in the form by which a three-dimensional model is set with respect to a virtual area. The bird's-eye view of the character arrange | positioned in the image generation space in a 2nd form. The figure which shows that four function areas are provided. The figure which shows the procedure which produces | generates the image of a character when seeing a character from arbitrary viewpoint positions. The figure which shows the hardware constitutions of the image processing system in a 2nd form. The figure which shows the data structure of the model data in a 2nd form. The figure which shows the data structure of function area information. The flowchart which shows the flow of the process in the rendering process in a 2nd form. The figure which shows that the reference | standard area is provided in addition to the function area. The figure which shows another form regarding the setting of the reference axis in a 2nd form. The figure which shows the relationship between a virtual area and a direction vector in the form by which a three-dimensional model is set with respect to a virtual area in a 2nd form.

(1) First mode (first image processing system)
First, an outline of the first image processing system of the present invention will be described with reference to FIGS. FIG. 1 shows an image generation space that is a virtual three-dimensional space in which the head of character A (hereinafter simply referred to as “character A”) centered on a center point CP is aligned with the Z-axis direction. It is an overhead view which shows the appearance arrange | positioned in GS. The nose portion An in the overhead view shows the front direction of the character A for convenience, and is not necessarily the actual nose portion of the character A. In the character A, a reference axis CA that is parallel to the Z-axis direction and passes through the center point CP of the character A is set. The center point CP is the midpoint of the character A in the vertical direction and the horizontal direction. The direction of the viewpoint position VP with respect to the character A is referred to as a viewpoint direction VD, and in this embodiment is the direction of the viewpoint position VP with respect to the reference axis CA. That is, the viewpoint direction VD of the present embodiment is a direction indicated by the direction vector Vd that is lowered from the viewpoint position VP perpendicularly to the reference axis CA. Since the viewpoint direction VD moves so as to rotate about the reference axis CA, the viewpoint direction VD moves so as to draw a virtual circle Cir on a plane (XY plane) perpendicular to the reference axis CA.

  In FIG. 1, a viewpoint direction VD1 is a viewpoint direction VD when the character A is viewed from the front. In this embodiment, the viewpoint direction VD1 is used as a reference direction for specifying the viewpoint direction VD for the character A. The viewpoint direction VD2 is a direction moved 45 degrees with respect to the circumferential direction CD with respect to the viewpoint direction VD1, and the viewpoint direction VD3 is a direction moved 90 degrees with respect to the circumferential direction CD with respect to the viewpoint direction VD1. For example, the viewpoint direction VD of the viewpoint position VP1 on the direction vector Vd1 indicating the viewpoint direction VD1 is the viewpoint direction VD1. Similarly, the viewpoint direction VD of the viewpoint position VP2 is the viewpoint direction VD2, and the viewpoint direction VD of the viewpoint position VP3 is the viewpoint direction VD3.

  Hereinafter, a relative distance between the viewpoint directions VD is indicated by an angle rotated with respect to the circumferential direction CD. In FIG. 1, the viewpoint positions VP1 to VP3 have the same distance from the reference axis CA, but the distance from each viewpoint position VP to the reference axis CA does not matter. The virtual circle Cir is a conceptual circle formed in the horizontal direction around the reference axis CA, and the length of the radius is not limited.

  2 to 4 show the case where the center point CP is viewed from each viewpoint position VP1, VP2, VP3 when the Z-axis values of the viewpoint positions VP1, VP2, VP3 are the same as the Z-axis value of the center point CP of the character A. This is an image of the character A to be expressed. The front character image A1 is an image of the character A to be expressed when the center point CP is viewed from the viewpoint position VP1, and the 45-degree character image A2 is to be expressed when the center point CP is viewed from the viewpoint position VP2. The 90-degree character image A3 is an image of the character A and should be expressed when the center point CP is viewed from the viewpoint position VP3.

  As shown in FIGS. 2 to 4, the hairstyle and mouth state of the character A are already displayed in the 45 degree character image A2 in the same manner as the state in the 90 degree character image A3. The eyes in the character image A2 are in a state where the eyes in the front character image A1 are just moved slightly to the left. As described above, the character A is not an aspect corresponding to the movement amount of the viewpoint direction VD, but has a feature that the two-dimensional expression is switched with respect to the specific viewpoint direction VD. For such an image of the character A, as in a conventional image processing system, one 3D model corresponding to the character A is generated, and the viewpoint position VP is moved relative to the generated 3D model to be used as a monitor. It cannot be expressed by the display method.

  In the first image processing system of the present invention, when it is desired to express a two-dimensional change according to the viewing angle, it corresponds to the direction in which the change is desired to be expressed (for example, the front direction, 45 degree direction, 90 degree direction). When three-dimensional models M1, M2, and M3 corresponding to the respective viewpoint directions VD1, VD2, and VD3 are created, and the character A viewed from the viewpoint position VP in each viewpoint direction VD1, VD2, and VD3 is expressed. Uses the corresponding three-dimensional model M. Hereinafter, the specific viewpoint direction VD in which the corresponding three-dimensional model M is prepared is sometimes referred to as “model viewpoint direction VDm”.

  On the other hand, the viewpoint direction VD for which the corresponding three-dimensional model M is not prepared, for example, as shown in FIG. 5, the character A corresponding to the interpolation viewpoint direction VDi between the model viewpoint direction VDm1 and the model viewpoint direction VDm2. The image is generated by interpolating the vertices of each three-dimensional model M by a predetermined interpolation method using the front model M1 corresponding to the model viewpoint direction VDm1 and the 45 degree model M2 corresponding to the model viewpoint direction VDm2. The Hereinafter, an image corresponding to the interpolation viewpoint direction VDi, that is, a two-dimensional image generated by interpolation using the two three-dimensional models M is referred to as an “interpolated image INima”.

  A range from one model viewpoint direction VDm to the next model viewpoint direction VDm with respect to the circumferential direction CD, that is, a range where the interpolation viewpoint direction VDi exists is hereinafter referred to as “interpolation range INr”. In this embodiment, the range from the model viewpoint direction VDm1 to the model viewpoint direction VDm2 is the interpolation range INr-1, and the range from the model viewpoint direction VDm2 to the model viewpoint direction VDm3 is the interpolation range INr-2. In the present embodiment, when the interpolation process is performed based on the two three-dimensional models M, the influence of each model viewpoint direction VDm depends on the distance between the interpolation viewpoint direction VDi and each model viewpoint direction VDm. The influence degree can be set as a parameter indicating.

  The influence degree will be described with reference to FIGS. 6 and 7. The influence degree is set so as to be obtained by a predetermined function according to the distance between the interpolation viewpoint direction VDi and each model viewpoint direction VDm, for example. Further, a desired influence degree may be set for a specific interpolation viewpoint direction VDi within the interpolation range INr. In this embodiment, as shown in FIG. 6, four interpolation viewpoint directions VDi1, VDi2, VDi3, and VDi4 are set in the interpolation range INr-1 as the interpolation viewpoint directions VDi for which the degree of influence is set. Interpolated viewpoint directions VDi1 to VDi4 are directions shifted by 9 degrees with respect to the circumferential direction CD from the model viewpoint direction VDm1. The degree of influence in this embodiment is a ratio given by each of the three-dimensional models M1 and M2 when the total degree of influence on the interpolation processing is 100%.

  For example, when it is desired to change the state of the character A uniformly from the front model M1 to the 45 degree model M2 according to the viewpoint direction VD, the influence degree of each of the three-dimensional models M1 and M2 is the influence degree table of FIG. It is set like T1. If the increase amount and decrease amount of the influence degree of each of the three-dimensional models M1 and M2 are set to be constant as in the influence degree table T1, the state of the character A can be changed uniformly. On the other hand, when the influence of either one of the front model M1 and the 45 degree model M2 is to be increased and changed, the influence degree of each of the three-dimensional models M1 and M2 is changed as shown in the influence degree table T2 of FIG. You only have to set it.

  In the influence degree table T2, the interpolation viewpoint direction VDi3 is closer to the model viewpoint direction VDm2 than the model viewpoint direction VDm1, but in the interpolation viewpoint direction VDi3, the influence degree of the front model M1 corresponding to the model viewpoint direction VDm1 is greater than the model viewpoint. It is set higher than the influence degree of the 45 degree model M2 corresponding to the direction VDm2. Thereby, in the interpolation range INr-1, it is possible to increase the influence of the front model M1 and generate the interpolated image INima.

  The influence degree of the interpolation viewpoint direction VDix that is not set in the influence degree table T may be set as appropriate according to the distance relationship between each of the two interpolation viewpoint directions VDi that sandwich the interpolation viewpoint direction VDix and the interpolation viewpoint direction VDix. . For example, in the case of the interpolation viewpoint direction VDix illustrated in FIG. 6, the degree of influence set for each of the interpolation viewpoint direction VDi1 and the interpolation viewpoint direction VDi2, the distance between the interpolation viewpoint direction VDi1 and the interpolation viewpoint direction VDix, and the interpolation What is necessary is just to set according to the relative relationship of the distance of viewpoint direction VDi2 and interpolation viewpoint direction VDix.

  A specific procedure for obtaining a character image based on the three-dimensional models M1, M2, and M3 when the predetermined gazing point GP is viewed from the predetermined viewpoint position VP will be described. In the following description, when the three-dimensional model M of the character A is arranged in the image generation space GS, as described above, the front direction of the three-dimensional model M is the reference direction VD1. The three-dimensional model M is arranged so that the vertical direction coincides with the Z-axis direction.

  First, as shown in FIG. 8A, when the viewpoint direction VD of the viewpoint position VP is the model viewpoint direction VDm1, the front model M1 is arranged in the image generation space GS as the three-dimensional model M of the character A. Then, the front model M1 is projected onto a plane P1 that includes the viewpoint position VP, and the gaze vector Vg from the viewpoint position VP toward the gazing point GP is orthogonal, and the projection image Ima-1 is used as a character image as a predetermined drawing area. Is generated. As shown in FIG. 8B, when the viewpoint direction VD of the viewpoint position VP is the model viewpoint direction VDm2, the 45-degree model M2 is arranged in the image generation space GS as the three-dimensional model M of the character A. Then, the 45 degree model M2 is projected onto the plane P2 including the viewpoint position VP and orthogonal to the gaze vector Vg, and the projection image Ima-2 is generated as a character image in a predetermined drawing area.

  Further, as shown in FIG. 8C, when the viewpoint direction VD corresponding to the direction vector Vd is the model viewpoint direction VDm3, the 90-degree model M3 is arranged as a three-dimensional model of the character A in the image generation space GS. Then, the 90-degree model M3 is projected on a plane P3 including the viewpoint position VP and orthogonal to the gaze vector Vg, and the projection image Ima-3 is generated as a character image in a predetermined drawing area. Projecting the three-dimensional model M onto the planes P1, P2, and P3 means so-called rendering of the three-dimensional model M on a predetermined plane.

  Next, a method for generating the interpolated image INima obtained based on the two three-dimensional models M will be described. The interpolation image INima generation method includes a first image generation method for generating an interpolation image INima by interpolating two-dimensional images obtained from two three-dimensional models M, and an interpolation process from two three-dimensional models M. There is a second image generation method for obtaining a three-dimensional interpolation model Mi and generating an interpolated image INima. For convenience of explanation, the interpolation image INima-i2 when the gazing point GP on the reference axis CA is viewed from the viewpoint position VP4 on the same virtual circle Cir as the viewpoint position VP1 and the viewpoint position VP2 in FIG. To do.

  Since the direction vector Vd4 of the viewpoint position VP4 coincides with the interpolation viewpoint direction VDi2, that is, the viewpoint direction VD of the viewpoint position VP4 is the interpolation viewpoint direction VDi2, the two three used to generate the interpolation image INima-i2 The dimension model M is a front model M1 and a 45-degree model M2 corresponding to the model viewpoint directions VDm1 and VDm2 adjacent to each other with the interpolation viewpoint direction VDi2 interposed therebetween.

  A procedure for generating the interpolated image INima-i2 by the first image generation method will be described with reference to FIG. First, the front model M1 is arranged in the image generation space GS, and a provisional projection image Ima-6 is obtained from the front model M1. The temporary projection image Ima-6 is in the model viewpoint direction VDm1, and is perpendicular to the gaze vector Vg1 from the viewpoint position VP1 on the virtual circle Cir passing through the viewpoint position VP4 to the gazing point GP, and includes the viewpoint position VP1. It is obtained by projecting the front model M1 onto the plane P6.

  Next, the 45 degree model M2 is arranged in the image generation space GS, and a provisional projection image Ima-7 is obtained from the 45 degree model M2. The temporary projection image Ima-7 is in the model viewpoint direction VDm2, and is perpendicular to the gaze vector Vg2 from the viewpoint position VP2 on the virtual circle Cir passing through the viewpoint position VP4 to the gazing point GP, and includes the viewpoint position VP2. It is obtained by projecting a 45 degree model M2 on the plane P7.

  In the provisional projection image Ima-6 and the provisional projection image Ima-7, for a non-matching part, a predetermined interpolation method is used by using the coordinates of the provisional projection images Ima-6 and Ima-7 corresponding to the non-matching part. Thus, the coordinates of the interpolated image INima-i2 corresponding to the viewpoint position VP4 are interpolated. As an interpolation method to be used, a known interpolation method such as Bezier interpolation or linear interpolation may be appropriately employed.

  The coordinates obtained by the interpolation processing are set as the coordinates of the portion of the interpolated image INima-i2 corresponding to the “non-matching portion”. The coordinates of the “matching part” are set directly as the coordinates of the interpolated image INima-i2 of the part corresponding to the “matching part”. In the interpolation processing, the influence degree table T in which the influence degree of each of the three-dimensional models M1 and M2 is set is referred to, and the process is performed with weighting each coordinate according to the set influence degree. By the above method, the interpolated image INima-i2 when the gazing point GP is viewed from the viewpoint position VP4 can be obtained.

  Next, a procedure for generating the interpolated image INima-i2 by the second image generation method will be described with reference to FIG. In the second image generation method, first, the three-dimensional interpolation model Mi corresponding to the viewpoint direction VDi2 is generated by interpolating the vertices of the front model M1 and the 45-degree model M2 by a predetermined interpolation method. The generated three-dimensional interpolation model Mi is arranged in the image generation space GS, and the three-dimensional interpolation model Mi is projected onto a plane Pi2 that is perpendicular to the gaze vector Vg4 from the viewpoint position VP4 to the gazing point GP and includes the viewpoint position VP4. The projection image is generated in a predetermined drawing area as an interpolation image INima-i2. As will be described later, normals are set at the vertices of the three-dimensional model M of the present embodiment. In the first image generation method and the second image generation method, interpolation processing is performed on the normals of the respective vertices by a conventional interpolation method as in the case of coordinates.

  Hereinafter, the image processing system 1 of the present invention that realizes the above configuration will be described. FIG. 11 is a diagram showing an example of the configuration of the image processing system 1 of the present invention. The image processing system 1 includes an operation input unit 2 that receives a user operation, a display unit 3 that presents a predetermined image to the user, a storage unit 4 that stores a computer program and various data for realizing the present invention, It is comprised with the control part 5 which controls operation | movement of the image processing system 1. FIG. The operation input unit 2 includes, for example, a keyboard, a mouse, a touch panel, and the like. The display unit 3 includes, for example, a monitor and a projector.

  The control unit 5 is composed of various storage areas such as a CPU, RAM and ROM necessary for its operation, and a drawing memory as a drawing area. The control unit 5 mainly executes interpolation by executing a computer program stored in the storage unit 4. It functions as an image generation means 5a, an interpolation image display means 5b, a cutout means 5c, a correction means 5d, a second model setting means 5e, and a viewpoint direction specifying means 5g. In addition to the computer program, the storage unit 4 stores model data MD, which is information related to the three-dimensional model M, and interpolation information INinf, which is information related to the interpolated image INima. Thereby, the storage unit 4 functions as a model data storage unit and an interpolation information storage unit.

  The model data MD is information regarding the three-dimensional model M corresponding to each model viewpoint direction VDm with respect to the character A. As shown in FIG. 12, the model data MD includes a corresponding viewpoint direction 10 and a plurality of parts information PI1, PI2,. Hereinafter, when it is not necessary to distinguish the parts information PI1, PI2,..., They are referred to as “part information PI”. The “parts” in the model data MD is a part in the three-dimensional model M, and is a group of vertices in which the manner of change due to movement of the viewpoint direction VD changes in the same way. Specific “parts” include, for example, hair, nose, mouth, and eyes. However, the “parts” in the present embodiment may be a plurality of distant parts or set as one “part” as long as the aspect of change due to movement in the viewpoint direction VD changes in the same way. Good.

  The part information PI is configured by associating vertex information 30 with part identification information 31 for identifying a part. The vertex information 30 is information regarding each vertex of a plurality of polygons constituting the part corresponding to the part identification information 31, and includes coordinate information and normal information of the coordinates of each vertex. The coordinate information may be a relative coordinate with respect to a predetermined point, or may be a coordinate value in the image generation space GS. The number of vertices of each three-dimensional model M prepared for the character A and the correspondence between the polygons are the same, and information indicating the correspondence between the vertices between the three-dimensional models is also included in the vertex information 30. .

  As shown in FIG. 13, the interpolation information INinf is composed of a plurality of range information RI1, RI2,. Hereinafter, the range information RI1, RI2,... Are referred to as “range information RI” when it is not necessary to distinguish them. In the range information RI, information related to the interpolation range INr is set, and includes an interpolation range 50, a use model 60, an interpolation method 70, and an influence degree 80. In the interpolation range 50, information for specifying the corresponding interpolation range INr is set. For example, in the case of the interpolation range INr-2, the model viewpoint direction VDm2 and the model viewpoint direction VDm3 that constitute the interpolation range INr-2 may be set.

  In the usage model 60, a three-dimensional model M corresponding to each of the two model viewpoint directions VDm constituting the corresponding interpolation range INr is set. In the interpolation method 70, an interpolation method applied to the corresponding interpolation range INr is set. In the influence degree 80, an influence degree table T indicating the influence degree of each three-dimensional model M set in the use model 60 is set. Further, part interpolation information 90 in which an interpolation method and an influence degree for a specific part are set with respect to the interpolation range RI is also included. Corresponding part identification information is associated with the part interpolation information. In FIG. 13, the range information RI1 includes part 1 interpolation information 90a in which an interpolation method for part 1 is set, and the range information RI2 includes part 2 interpolation information 90b in which an interpolation method for part 2 is set. ing. Thereby, the memory | storage part 4 functions as a group interpolation information storage means.

  A modeling process for generating the model data MD and the interpolation information INinf regarding the character A will be described with reference to the flowchart of FIG. In this embodiment, an interpolation range INr-1 from the model viewpoint direction VDm1 to the model viewpoint direction VDm2 will be mainly described. The modeling process is controlled by the control unit 5. First, information for specifying an interpolation range is set in the interpolation range 50 in step S100. In this embodiment, information specifying the interpolation range INr-1 is set. In this embodiment, information for specifying a range from the model viewpoint direction VDm1 to the model viewpoint direction VDm2 is set. The setting is set by a user input operation, for example.

  Next, in step S105, a three-dimensional model M as a first three-dimensional model is set. When the 3D model M to be set has not yet been generated as at the start of the modeling process, it is generated by a known 3D model generation method. For example, when the first three-dimensional model M is the front model M1, by generating the front model M1, model data MD1 of the front model M1 is generated. The part information PI and the vertex information 30 are associated with each other by, for example, specifying a part of the generated front model M1 by dragging the mouse and identifying the part with respect to the specified part. Information 31 may be set.

  Next, in step S110, based on the three-dimensional model M set in step S100, the model viewpoint position VDm constituting the interpolation range INr-1 set in step S100 is set as the second three-dimensional model. A corresponding three-dimensional model M (45 degree model M2 in this embodiment) is generated. In the case of this embodiment, the front model M1 as the first three-dimensional model is displayed on the monitor 3, and the front model M1 is rotated in a state viewed from the model viewpoint direction VDm2. In the front model M1 in this state, an area having a state different from that of the 45 degree model M2 is cut out, the cut out area is transformed into the state of the 45 degree model M2, and the deformed area is displayed on the front surface displayed on the monitor 3. Return to model M1.

  In the front model M1, the 45 ° model M2 corresponding to the model viewpoint direction VDm2 is finally generated by performing the above processing on all regions having a state different from that of the 45 ° model M2. That is, the model data MD2 of the 45 degree model M2 is generated. By the processing in step S110, the control unit 5 functions as a cutout unit 5c, a correction unit 5d, and a second model setting unit 5e. Note that when the cut-out area is deformed, the correspondence between a plurality of polygons constituting the area does not change before and after the deformation. Therefore, each vertex of the polygon constituting the front model M1 corresponds to each vertex of the polygon constituting the 45 degree model M2. In each of the vertex information 30 of the model data MD1 and the vertex information 30 of the model data MD2, for example, a corresponding vertex is set for each vertex so that the corresponding vertex is identified.

  Next, proceeding to step S120, range information RI corresponding to the interpolation range INr-1 is generated by initial setting. In this embodiment, the range information RI is initialized as follows. The interpolation range set in step S100 is set in the interpolation range 50, and a predetermined interpolation method and a predetermined influence degree table T are set in advance for each of the interpolation method 70 and the influence degree 80. In the usage model 60, model data MD1 and MD2 corresponding to the model viewpoint directions VDm1 and VDm2 constituting the interpolation range INr-1 are set with reference to the model data MD.

  When the initial setting range information RI is generated, the process proceeds to step S125, and display processing is performed. In the display process, when the viewpoint position VP corresponding to the range information RI is determined in the image generation space GS, the character A when the center point CP is the gazing point GP, that is, when the center point CP is viewed from the viewpoint position VP. These images are generated as described above and displayed on the monitor 3 by being drawn in the drawing memory. In the present embodiment, the interpolated image INima corresponding to the interpolated viewpoint direction VDi between the model viewpoint direction VDm1 and the model viewpoint direction VDm2 is described above based on the model data MD1 of the front model M1 and the model data MD2 of the 45 degree model M2. It is obtained by performing an interpolation process (first image generation method or second image generation method). The interpolation viewpoint direction VDi is specified as the direction of the viewpoint position VP with respect to the reference axis CA. Thereby, the control part 5 functions as the interpolation image generation means 5a, the interpolation image display means 5b, and the interpolation viewpoint direction specification means 5g.

  Then, after a series of images of the character A is displayed on the monitor 3, it is determined whether or not the range information RI is corrected in step S130. If there is a predetermined correction operation by the user, the process proceeds to step S135 to enter a state for accepting correction of the range information RI. For example, when it is desired to relatively increase the influence of the front model M1 regarding the generation of the interpolated image INima, the interpolation viewpoint in which the influence degree of the front model M1 is larger than the influence degree of the 45 degree model M2 as in the influence degree table T2. What is necessary is just to set so that the range of direction VDi may become wide. If there is a more suitable interpolation method, the interpolation method may be changed to a more suitable interpolation method.

  Moreover, you may comprise so that the interpolation method and influence degree can be set for every part. In that case, when the part is designated by the user, the part interpolation information 90 in the range information RI being processed is set for the designated part. As described above, in order to obtain an image of the desired character A, after the range information RI is corrected, the process returns to step S125 and the display process is performed. Thereby, the state after correction can be confirmed. When the part interpolation information 90 is set, an interpolation process is performed on the corresponding part based on the interpolation method and the degree of influence set in the part interpolation information 90.

  When there is no correction operation in step S130, it is determined that the range information RI has been generated and the process proceeds to step S140. In step S140, it is determined whether the user has performed an end operation or a continuation operation. If the continuation operation has been performed, the process returns to step S100. Thereby, the range information RI corresponding to the next interpolation range INr can be generated. For example, when generating the range information RI corresponding to the interpolation range INr-2 from the model viewpoint direction VDm2 to the model viewpoint direction VDm3, the interpolation range is set in step S100. Thereafter, processing similar to the processing related to the interpolation range INr-1 is performed, and range information RI related to each interpolation range is generated.

  Note that the first three-dimensional model set in step S105 does not have to be the three-dimensional model M used in the interpolation of the range information RI to be generated, but the second three-dimensional model generated in step S110. What is necessary is just to set the three-dimensional model M suitably so that the production | generation of the three-dimensional model M becomes easy. For example, the front model M1 may always be set as the first three-dimensional model M.

  If it is determined in step S140 that the user has performed an end operation, the storage process is performed in step S145, and then the modeling process is terminated. In the storage process, all range information RI generated in the modeling process is stored in the storage unit 4 as interpolation information INinf regarding the reference axis CA of the character A.

  Next, a rendering process for displaying the character A of the application on the monitor 3 during execution of the predetermined application will be described. First, the case where rendering processing is performed based on the second image generation method will be described with reference to the flowchart shown in FIG. The rendering process is controlled by the control unit 5. In this embodiment, a case where the application in which the character A is displayed on the monitor 3 is a game will be described. In this case, by providing the control unit 5 with a game control unit that controls execution of a predetermined game, the image processing system 1 functions as a game system, and the character A is displayed on the game screen during the game.

  First, in step S200, the viewpoint position VP is specified in the image generation space GS in which the character A is arranged. The position where the character A is arranged is specified by the center point CP, for example. The position where the character A is arranged and the viewpoint position VP may be determined according to the user's operation content, game content, score, and the like. In the following rendering processing, as shown in FIG. 16, a case where an image displayed on the monitor 3 is generated when the gazing point GP is viewed from the specified viewpoint position VP9 will be described. In step S205, the viewpoint direction VD of the viewpoint position VP9 is specified. Thereby, the control part 5 functions as the viewpoint direction specifying means 5g. The viewpoint direction VD of the viewpoint position VP9 is determined by the direction vector Vd9 of the viewpoint position VP9. In this embodiment, the direction vector Vd9 indicates the interpolation viewpoint direction Di2. That is, the viewpoint direction VD of the viewpoint position VP9 is the interpolation viewpoint direction VDi2.

  After the determination of the viewpoint direction VD, in step S210, the range information RI corresponding to the interpolation viewpoint direction VDi2 is selected with reference to the interpolation information INinf. With reference to the interpolation range 50 of each range information RI in the interpolation information INinf, the interpolation range 50 including the interpolation viewpoint direction VDi2 is specified, and the range information RI corresponding to the specified interpolation range 50 is set in the interpolation viewpoint direction VDi2. The corresponding range information RI is selected. For example, the interpolation range 50 including the interpolation viewpoint section VDi2 may be specified based on the distance from the reference direction V1. In this embodiment, the range information RI in which the interpolation range INr-1 from the model viewpoint direction VDm1 to the model viewpoint direction VDm2 including the interpolation viewpoint direction VDi2 is set as the interpolation range 50 is selected. Thereby, two model viewpoint directions VDm1 and VDm2 with respect to the interpolation viewpoint direction VDi2 are determined, and the control unit 5 functions as a model viewpoint direction setting unit.

  Subsequently, the process proceeds to step S220, and the two three-dimensional models M used for the interpolation process are specified with reference to the use model 60 in the selected range information RI. In this embodiment, the front model M1 and the 45 degree model M2 are specified. Next, in step S230, the identified front model M1 and 45 degree model M2 are placed in the image generation space GS, and the above-described interpolation processing is performed to generate the three-dimensional interpolation model Mi. Thereby, the control part 5 functions as an interpolation model production | generation means. The interpolation processing is performed based on the interpolation method 70, the influence degree 80, and the part interpolation method 90 with reference to the selected range information RI. The influence degree is specified by referring to the influence degree table T set to the influence degree 80. When the interpolation viewpoint direction VDix is not set in the influence degree table T, it is set in the influence degree table T, two interpolation viewpoint directions VDi sandwiching the interpolation viewpoint direction VDix are specified, and the circumferential direction with each interpolation viewpoint direction VDi What is necessary is just to specify the influence degree of each three-dimensional model M with respect to the interpolation viewpoint direction VDix according to the distance regarding CD. Thereby, the control part 5 functions as an influence degree specifying means.

  Next, in step S240, the generated three-dimensional interpolation model Mi is arranged in the image generation space GS and is three-dimensionally displayed on a plane P9 that includes the viewpoint position VP and is orthogonal to the gaze vector Vg9 from the viewpoint position VP9 toward the gazing point GP. The interpolation model Mi is projected. Thereby, the interpolation image INima-i2 is obtained on the plane P9. In step S250, the interpolated image INima-i2 is drawn in the drawing memory, so that the interpolated image INima-i2 of the character A is displayed on the monitor 3. This completes the rendering process.

  If the first image generation method is applied to the rendering process, provisional projection images Ima-10 and Ima-11 corresponding to the model viewpoint directions VDm1 and VDm2 are generated in step S230, and in step S240. An interpolation image INima-i2 is generated on the plane P9 by performing interpolation processing of the two provisional projection images Ima-10 and Ima-11, and the generated interpolation image INima-i2 is drawn in step S250. As a result, the interpolated image INima-i2 is displayed on the monitor 3. The processing from step S200 to step S220 is the same as the processing in the case of the second image generation method.

  A specific procedure when the first image generation method is applied will be described with reference to FIG. In step S230, the following processing is performed to obtain provisional projection images Ima-10 and Ima-11. First, a temporary viewpoint position VP10 that is in the model viewpoint direction VDm1 and is on the virtual circle Cir that passes through the viewpoint position VP9 is specified, and a temporary viewpoint position VP11 that is in the model viewpoint direction Dm2 and is on the virtual circle Cir that passes through the viewpoint position VP. Is identified. The front model M1 arranged in the image generation space GS is projected onto a plane P10 that is orthogonal to the gaze vector Vg10 viewed from the temporary viewpoint position VP10 and viewed from the temporary viewpoint position VP10 and includes the temporary viewpoint position VP10. A projection image Ima-10 is generated. Thereby, the control unit 5 functions as a first temporary image acquisition unit.

  Further, the 45-degree model M2 that is orthogonal to the gaze vector Vg11 from the temporary viewpoint position VP11 to the gazing point GP and that is disposed in the image generation space GS is projected onto the plane P11 that includes the temporary viewpoint position VP11. A temporary projection image Ima-11 is generated. Thereby, the control part 5 functions as a 2nd temporary image acquisition means. Next, in step S240, by performing the above-described interpolation processing based on the temporary projection image Ima-10 and the temporary projection image Ima-11, the interpolation image INima-i2 is orthogonal to the gaze vector Vg9, and the viewpoint position It produces | generates on the plane P9 containing VP9.

  The present invention is not limited to the above-described embodiment, and may be implemented in various forms. For example, the position and number of the model viewpoint direction VDm set on the virtual circle Cir may be set as appropriate in accordance with the characteristics to be displayed by the character A. When the interpolation method and the influence degree are constant regardless of the interpolation range INr, it is not necessary to set the range information RI regarding each interpolation range INr. The reference direction for specifying the viewpoint direction VD is not limited to the front direction, and any viewpoint direction VD may be used as the reference direction. Each vertex coordinate of the three-dimensional model M corresponding to the model viewpoint direction VDm may be obtained by a function based on each vertex of one basic three-dimensional model. In this case, a function for obtaining vertex coordinates is set in the vertex information 30 of the model data MD.

  When the image processing system 1 functions only as a game system, the control unit 5 may not have a function for performing modeling processing such as the cutout unit 5a, the correction unit 5d, and the second model setting unit. . The same processing may be applied to the three-dimensional model M including not only the head of the character A but also the body part. The target displayed on the monitor based on the three-dimensional model M is not limited to a character, and may be a living thing, a plant, or an inorganic substance.

  Further, the reference axis CA may be an axis that passes through the character A, and may be set as appropriate according to the characteristics of the character A. In this case, the virtual circle Cir formed by the viewpoint direction VD is formed on a plane perpendicular to the reference axis CA with the reference axis CA at the center.

  For example, as shown in FIG. 18, a reference axis CAa parallel to the X-axis direction perpendicular to the YZ plane may be set. A character A that changes two-dimensionally in the vertical direction can be handled. In this case, the reference axis CAa is the Z axis of the above form, the virtual circle Cira formed on the YZ plane perpendicular to the reference axis CAa is the virtual circle Cir in the above form, and the viewpoint direction VD indicating the characteristic aspect is For example, the model viewpoint directions VDm10, VDm11, and VDm12 may be set, and the same processing as in the above embodiment may be performed. Further, a reference axis CAb that is not parallel to any of the X axis, the Y axis, and the Z axis may be set. In this case, the same processing as in the above-described embodiment may be performed with the reference axis CAb as the Z-axis in the above form and the virtual circle Cirb formed on the plane perpendicular to the reference axis CAb as the virtual circle Cir in the above form.

  Further, the present invention may be realized in units of a virtual area VA including a plurality of viewpoint directions VD. In this case, as shown in FIG. 19, a virtual sphere G centered on the center point CP of the character A is arranged in the image generation space GS. Then, a plurality of virtual areas VA are formed by the horizontal virtual latitude VL and the vertical virtual meridian VM, and specific model areas VAm1 and VAm2 of the virtual area VA are drawn in a direction of drawing a circle around the center point CP. And a three-dimensional model M corresponding to each model region VAm is associated.

  In this case, the direction vector Vd for determining the direction of the character A with respect to the viewpoint position VP is a vector from the viewpoint position VP toward the center point CP of the character A. When passing through the character area VAm like the direction vector Vd100, the three-dimensional model M corresponding to the model area VAm through which the direction vector Vd passes is arranged in the image generation section GS to generate an image corresponding to the viewpoint position VP. When passing through the interpolation area VAi that is not the model area VAm as in the direction vector Vd110, for example, two model areas VAm1 and VAm2 sandwiching the interpolation area VAi are specified, and 3 corresponding to the specified model areas VAm1 and Rm2 are specified. The interpolation image INima corresponding to the interpolation area VAi may be generated by performing the interpolation processing as described above using the dimension model M.

  The model area VAm used for the interpolation processing related to the specific interpolation area VAi may be set in advance for each interpolation area VAi, for example. The plurality of virtual areas VA set in the virtual sphere G may be arranged in the circumferential direction with the center point CP as the center, and the shape and size are not limited.

(2) Second form (second image processing system)
The outline of the second image processing system of the present invention will be described with reference to FIGS. Similar to the first image processing system, the second image processing system cannot represent a character by using a conventional method of representing a character by viewing one three-dimensional model from an arbitrary viewpoint position. Dimensional expression is possible. Hereinafter, differences from the first image processing system will be mainly described. The concepts of the center point CP, the reference axis CA, the viewpoint positions VP1 to VP3, the direction vectors Vd1 to Vd3, and the viewpoint directions VD1 to VD3 are the same as in the first embodiment. In this embodiment, the viewpoint direction VD20 is the viewpoint direction VD moved 135 degrees from the viewpoint direction VD1, the viewpoint position VP20 is the viewpoint position VP on the viewpoint direction VD20, and the direction vector Vd20 is the viewpoint position VP20. To the vector perpendicular to the reference axis CA.

  Hereinafter, in this embodiment, the viewpoint direction VD1 is referred to as a reference direction VDc, and the other viewpoint directions VD, that is, the viewpoint direction VD moved from the reference direction VDc to the circumferential direction CD is referred to as a moving viewpoint direction MVD. For example, the viewpoint directions VD2, VD3, and VD20 are also moving viewpoint directions MVD2, MVD3, and MVD20 that have moved 45 degrees, 90 degrees, and 135 degrees from the reference direction VDc, respectively. In this embodiment, a three-dimensional model M used when the character A is viewed from the reference direction VDc is prepared as the reference model Mc, and the three-dimensional model M used when the character A is viewed from the moving viewpoint direction MVD is It is generated by the function F corresponding to the viewpoint direction MVD. Hereinafter, the three-dimensional model M used when the character A is viewed from the moving viewpoint direction MVD is referred to as “moving model Mm”.

  In the present embodiment, a plurality of function sections FR are provided for the circumferential direction CD centered on the center point CP, and each function section FR has a function F used for the moving viewpoint direction MVD included in the function section FR. Are associated. In the example of FIG. 21, for the character A, a function section FR1 of the reference direction VDc to the moving viewpoint direction MVD2, a function section FR2 of the moving viewpoint direction MVD2 to the moving viewpoint direction MVD3, and a function of the moving viewpoint direction MVD3 to the moving viewpoint direction MVD20. A section FR3 and a function section FR4 from the moving viewpoint direction MVD20 to the reference direction VDc are provided. In the present embodiment, each of the function sections FR1 to FR4 is associated with, for example, each of the functions F1 to F4.

  The function F1 is a function for generating a movement model Mm1 that is a three-dimensional model M of the character A corresponding to the movement viewpoint direction MVDfr1 in the function section FR1. Specifically, the function F1 sets each vertex coordinate of the reference model Mc and a movement amount Mq1 from the reference direction VDc to the movement viewpoint direction MVDfr1, so that each vertex coordinate of the movement model Mm1 is obtained. Is set.

  The function F2 is a function for generating a movement model Mm2 that is a three-dimensional model M of the character A corresponding to the movement viewpoint direction MVDfr2 in the function section FR2. Specifically, the function F2 sets each vertex coordinate of the reference model Mc and a movement amount Mq2 from the reference direction VDc to the movement viewpoint direction MVDfr2, so that each vertex coordinate of the movement model Mm2 is obtained. Is set.

  The function F3 is a function for generating a movement model Mm3 that is a three-dimensional model M of the character A corresponding to the movement viewpoint direction MVDfr3 in the function section FR3. Specifically, the function F3 sets each vertex coordinate of the reference model Mc and a movement amount Mq3 from the reference direction VDc to the movement viewpoint direction MVDfr3 so that each vertex coordinate of the movement model Mm3 can be obtained. Is set.

  Finally, the function F4 is a function for generating a movement model Mm4 that is a three-dimensional model M of the character A corresponding to the movement viewpoint direction MVDfr4 in the function section FR4. Specifically, the function F4 sets each vertex coordinate of the reference model Mc and a movement amount Mq4 from the reference direction VDc to the moving viewpoint direction MVDfr4 so that each vertex coordinate of the movement model Mm4 can be obtained. Is set. In addition, when the aspect which changes with parts of the character A differs, each function section FR is matched with the function F corresponding to each part.

  Next, as illustrated in FIG. 22, a state in which the gazing point GP is viewed from the viewpoint position VPa on the moving viewpoint direction MVDa with respect to the character A during execution of an application (for example, a game) in which the character A appears. The procedure for expressing the character A when displaying on the monitor will be described. As shown in FIG. 22, the moving viewpoint direction MVDa is included in the function section FR2, and therefore the function F2 associated with the function section FR2 is determined as the function F used for generating the moving model Mm.

  Each vertex coordinate of the movement model Mm is obtained by setting the movement amount Mqa in the movement viewpoint direction MVDa and each vertex coordinate of the reference model Mc to the function F2. As a result, a movement model Mm, which is a three-dimensional model M corresponding to the movement viewpoint position MVDa, is generated at a position centered on the center point CP. Subsequently, a projection image obtained by projecting the movement model Mm is drawn on a plane Pa orthogonal to the line-of-sight vector Vga from the viewpoint position VPa toward the gazing point GP. The projected image is drawn on a predetermined drawing area, and is displayed on the monitor as an image of the character A when the gazing point GP is viewed from the viewpoint position VPa.

  A specific configuration of the second image processing system 100 that realizes the above-described image display will be described. First, the hardware structure of the image processing system 100 is shown in FIG. The second image processing system 100 includes an operation input unit 2, a display unit 3, a storage unit 104, and a control unit 105. The control unit 105 includes a CPU and various storage areas such as RAM and ROM necessary for its operation. By executing the computer program in the storage unit 104, it mainly functions as the movement model generation unit 105a, the drawing unit 105b, the drawing display unit 105c, and the viewpoint direction specifying unit 105d. The function of each means will be described later.

  In addition to the computer program for realizing the present invention, the storage unit 104 stores model data MD ′ as reference model data for generating the reference model Mc in the image generation space GS, and function interval information FRI. Yes. The data structure of the model data MD ′ in this embodiment is shown in FIG. The model data MD ′ is composed of part identification information 31 and vertex information 30 as in the data structure of the model data MD of the first form. The vertex information 30 includes vertex coordinates and normal information of a plurality of polygons included in each part. The vertex coordinates may be relative coordinates with respect to a predetermined point, or may be coordinate values in the image generation space GS. In the second embodiment, the model data MD ′ is only the three-dimensional model M used when the character A is viewed from the reference direction VDc. Therefore, it is not necessary to associate the viewpoint direction VD with the model data MD ′. The storage unit 104 functions as a model data storage unit by storing the model data MD ′.

  Next, the data structure of the function section information FRI will be described with reference to FIG. The function section information FRI includes a function section FR, part information 120, and function information 130. In the part information 120, part identification information indicating at least one part in the reference model Mc is set, and in the function information 130, as a model forming function used for the corresponding part with respect to the corresponding function section FR. A function F and a condition related to the function F are set. The function F is a vertex of each part of the movement model Mc corresponding to the movement viewpoint direction MVD based on the movement amount Mq of the movement viewpoint direction MVD in the corresponding function section FR and each vertex coordinate of each part of the reference model Mc. It is set so that coordinates can be obtained. Thereby, when the character A viewed from an arbitrary angle is viewed, the desired character A can be expressed. The storage unit 104 functions as a function storage unit by storing the function F.

  The number of function information 130 associated with each function section FR varies depending on the function section FR. In the example of FIG. 25, among the four function sections FR1 to FR4, the function section FR1 is associated with three function information 130a to 130c, and each function information 130a to 130c is a part in which the function is used. Is associated with the part information 120a to 120c set. In addition, regarding the function section FR3, one function information 130g is associated, and the function information 130g is associated with part information 120g in which all parts are set. Thereby, the function F set in the function information 130g is applied to all parts.

  A rendering process for displaying the image of the character A on the monitor when the gazing point GP is viewed from an arbitrary viewpoint position VPa in the image generation space GS will be described with reference to the flowchart of FIG. For example, when the character A appears in the game by the rendering process, the character A viewed from the viewpoint position VPa is displayed on the monitor during the execution of the game. First, in step S500, the position of the character A in the image generation space GS is determined. For example, it may be determined by the position of the center point CP of the character A. Thereby, the coordinate value of the reference model Mc of the character A is determined. The position of the character A is determined by the situation of the game being executed and the operation of the player. Subsequently, the viewpoint position VPa is determined in step S505. The viewpoint position VPa is determined by the situation of the game being executed and the operation of the player.

  Next, in step S510, the moving viewpoint direction MVDa related to the viewpoint position VPa, that is, the viewpoint direction VD that is the direction of the character A with respect to the viewpoint position VPa is determined. Thereby, the control unit 105 functions as the viewpoint direction specifying unit 105d. The moving viewpoint direction MVD is determined as the direction indicated by the direction vector Vda perpendicular to the reference axis CA of the character A from the viewpoint position VPa. Next, in step S520, the movement amount Mqa is determined. The movement amount Mqa is an angle at which the reference direction VDc moves in the movement viewpoint direction MVD. After determining the movement amount Mqa, at least one function F is determined in step S530. The function section FR including the moving viewpoint direction MVD may be determined, and the function section information FRI related to the determined function section FR may be determined.

  After determining the function F to be used, the process proceeds to step S540 to generate a movement model Mm. The movement model Mm is generated based on the function section information FRI determined in step S530. For example, in step S530, when the function section information FRI of the function section FR1 is determined by referring to the function section information FRI in FIG. 25, the vertex coordinates and the movement amount Mqa of the reference model Mc are obtained from the function information 130a. , 130b, 130c, the vertex coordinates of the parts of the corresponding parts information 120a, 120b, 120c are obtained. Thereby, each vertex of the movement model Mm is obtained, and the movement model Mm is generated at the position determined in step S500.

  After the generation of the movement model Mm, the process proceeds to step S550. In step S550, the generated movement model Mm is projected onto a plane Pa that intersects perpendicularly to the line-of-sight vector Vga from the viewpoint position VP to the gazing point GP, and the projection image is drawn in a predetermined drawing area. Subsequently, in step S560, the projected image drawn in the drawing area is displayed on the monitor as an image of the character A.

  Through steps S510 to S540, the control unit 105 functions as the movement model generation unit 105a. In particular, in step S510, the control unit 105 functions as the viewpoint direction specifying unit 105d. Further, in step S550, the control unit 105 functions as the drawing unit 105b, and in step S560, the control unit 105 functions as the drawing display unit 105c.

  The present invention is not limited to the above-described form and may be executed in various forms. For example, the number of function sections FR set for the character to be displayed may be set according to the display mode. Further, a function F that is used only in a specific moving viewpoint direction MVD may be set. In that case, the specific moving viewpoint direction may be set in the function section of the function section information FRI. The model data MD ′ of the character A may be set with a function for obtaining the vertex coordinates of each polygon. The same processing may be applied to the three-dimensional model M including not only the head of the character A but also the body part. The target displayed on the monitor based on the three-dimensional model M is not limited to a character, and may be a living thing, a plant, or an inorganic substance.

  In addition to the function section FR in which the three-dimensional model M generated by the function F is used, a reference section CR in which the reference model Mc is used may be provided. An example in which two reference intervals CR1 and CR2 are provided is shown in FIG. Furthermore, the reference axis CA may be set as a reference axis CAa parallel to the X-axis direction perpendicular to the YZ plane, as shown in FIG. Thereby, it is possible to deal with the character A that changes two-dimensionally in the vertical direction. In this case, the reference axis CAa is the Z axis in the above form, and the virtual circle Cira formed on the YZ plane perpendicular to the reference axis CAa is the virtual circle Cir in the above form, and the character A used in the reference direction VDc100. The three-dimensional model M is set as the reference model Mc, the function section FR is set for the virtual circle Cira, and the same processing as in the above embodiment may be performed. Further, a reference axis CAb that is not parallel to any of the X axis, the Y axis, and the Z axis may be set. In this case, the same processing as in the above-described embodiment may be performed with the reference axis CAb as the Z-axis in the above form and the virtual circle Cirb formed on the plane perpendicular to the reference axis CAb as the virtual circle Cir in the above form.

  Further, the present invention may be realized in units of a virtual area VA including a plurality of viewpoint directions VD. In this case, as shown in FIG. 29, a virtual sphere G centered on the center point CP of the character A is arranged in the image generation space GS. A plurality of virtual areas VA are formed by the horizontal virtual latitude VL and the vertical virtual meridian VM. In this case, the direction vector Vd is a vector from the viewpoint position VP toward the center point CP of the character A. Like the direction vector Vd100, the reference model Mc is used for the moving viewpoint direction MVD corresponding to the direction vector Vd passing through the reference area VAc in the virtual area VA.

  On the other hand, a plurality of function section areas FRA are formed by one or more virtual areas VA, and a function F for generating the movement model Mm of the character A from the reference model Mc is set for each function section area FRA. . FIG. 29 shows a state in which the function section area FRA is set by eight virtual areas VA. In this case, the function F associated with the function section area F is a function F used for the moving viewpoint direction MVD corresponding to the direction vector Vd passing through the function section area FRA, like the direction vector Vd110. .

  The movement model Mm is generated by setting the movement amount between the reference direction Vc and the movement viewpoint direction MVD passing through the reference area Ac and the vertex coordinates of the reference model Mc to the function F. The plurality of virtual areas VA set in the virtual sphere G may be arranged in the circumferential direction with the center point CP as the center, and the shape and size are not limited.

A character 1 first image processing system 100 second image processing system 2 operation input unit 3 display unit 4, 104 storage unit 5, 105 control unit 5a interpolation image generation unit 5b interpolation image display unit CA reference axis CP center point Vd Direction vector VD View direction VDc Reference direction VDi Interpolated view direction VDm Model view direction MVD Moving view direction Mq Movement amount VP View position

Claims (18)

Based on the three-dimensional model of the display object placed at a predetermined position in the virtual three-dimensional space, the image of the display object is displayed on the display unit according to the viewpoint direction that is the direction of the viewpoint position with respect to the display object. An image processing system,
First model data for forming a first model used as a three-dimensional model of the display object when the viewpoint position is in the first viewpoint direction with respect to the display object, and the display Used as a three-dimensional model of the display object when the viewpoint position with respect to the object is in the second viewpoint direction in which the first viewpoint direction is moved in a predetermined circumferential direction around the display object Model data storage means for storing second model data for forming a second model to be generated;
An interpolation image that is a two-dimensional image of the display object corresponding to an interpolation viewpoint direction existing between the first viewpoint direction and the second viewpoint direction with respect to the circumferential direction is obtained from the first model data. Generated by performing an interpolation process based on each vertex coordinate of the obtained first model and each vertex coordinate of the second model obtained from the second model data, Interpolated image generating means for drawing in the drawing area;
Interpolation for displaying the interpolated image drawn in the drawing area on the display unit as the display object viewed from the viewpoint position when the viewpoint position is in the interpolation viewpoint direction with respect to the display object And an image display means. Each of the first and second model data includes association information that associates each vertex forming the first model with each vertex forming the second model,
The image processing system according to claim 1, wherein the interpolation image generation unit generates the interpolation image by performing an interpolation process while maintaining a correspondence relationship between the vertices. In each of the first and second model data, group identification information for identifying a corresponding group is associated with each vertex of the plurality of polygons so that each vertex forms a group,
A group interpolation information storage unit that stores an interpolation method used in the interpolation image generation unit in association with each group is further provided.
The interpolation image generation means generates the interpolation image by performing the interpolation processing based on each vertex coordinate corresponding to the group by an interpolation method associated with the group. The image processing system according to claim 2. Relative positions of the interpolation viewpoint direction set in the interpolation range between the first viewpoint direction and the second viewpoint direction, and each of the first viewpoint direction and the second viewpoint direction According to the relationship, the first model data and the second model data specify the first and second influence levels indicating the magnitude of the influence on the interpolation in the interpolated image generating means. Further having an influence degree specifying means for
The interpolation image generation means includes
Each of the vertex coordinates obtained from the first model data and each of the second model data obtained from the first model data on the basis of the first influence degree and the second influence degree specified by the influence degree specifying means. The image processing system according to claim 1, wherein interpolation processing based on vertex coordinates is performed. Further comprising an operation input means for receiving the user's input operation;
Cutout means for displaying the first model on the display unit, and cutting out a specific part of the first model displayed on the display unit in accordance with the operation of the user;
Correction means for accepting a correction operation by the user with respect to the cut out specific part;
2. The image processing system according to claim 1, further comprising: a second model setting unit that sets the first model in which the specific portion is corrected by the correcting unit as the second model. . A reference axis extending in a predetermined direction through the display object is set,
In the model data storage unit, the second viewpoint direction is a direction in which the first viewpoint direction moves in a predetermined circumferential direction around the reference axis,
The interpolation image generation means includes
Viewpoint direction specifying means for specifying the direction of the viewpoint position with respect to the reference axis as the viewpoint direction with respect to the display object is provided;
When the specified viewpoint direction exists between the first viewpoint direction and the second viewpoint direction, the interpolation image is generated with the viewpoint direction as the interpolation viewpoint direction. The image processing system according to claim 1. The model data storage means is a model for forming a three-dimensional model of the display object corresponding to each model viewpoint direction with respect to a plurality of model viewpoint directions set on a circumferential direction around the reference axis. Remember the data,
The interpolation image generation means includes
When the viewpoint direction specified by the viewpoint direction specifying unit is the interpolation viewpoint direction, two model viewpoint directions sandwiched with respect to the circumferential direction are specified, and one model viewpoint direction is set as the first viewpoint direction. And a model viewpoint direction setting means for setting the other model viewpoint direction as the second viewpoint direction,
The interpolated image generating means generates the interpolated image corresponding to the specified viewpoint direction based on the first viewpoint direction and the second viewpoint direction set by the model viewpoint direction setting means. The image processing system according to claim 6. Interpolation information storage means for storing range information in which information related to interpolation of two corresponding models is stored for each interpolation range constituted by the two model viewpoint directions adjacent to each other in the circumferential direction,
The model viewpoint direction specifying means specifies the two model viewpoint directions by specifying range information of an interpolation range including the specified viewpoint direction with reference to the interpolation information storage means,
The image processing system according to claim 7, wherein the interpolation image generation unit performs interpolation of model data corresponding to the two models based on information related to the interpolation. The interpolation image generation means includes
An interpolation model generating means for generating a three-dimensional interpolation model by interpolating each vertex coordinate obtained from the first model data and each vertex coordinate obtained from the second model data;
An image obtained from the three-dimensional interpolation model when the predetermined three-dimensional interpolation model is arranged at the predetermined position as the three-dimensional model of the display target object and a predetermined gaze point is viewed from the viewpoint position is displayed on the display target object The image processing system according to claim 7, wherein the image processing system is generated as the interpolated image. The interpolation image generation means includes
A point having the same distance from the viewpoint position and the reference axis and in the set first viewpoint direction is set as a first temporary viewpoint position, and the first corresponding to the set first viewpoint direction is set. A first temporary image obtained from the first model when a predetermined model is placed at the predetermined position as a three-dimensional model of the display object and the predetermined gazing point is viewed from the first temporary viewpoint position. First provisional image acquisition means for acquiring
A point having the same distance from the viewpoint position and the reference axis and in the set second viewpoint direction is set as a second temporary viewpoint position, and the second corresponding to the set second viewpoint direction is set. When the second model is placed at the predetermined position as a three-dimensional model of the display object and the gazing point is viewed from the second temporary viewpoint position, a second temporary image obtained from the second model is acquired. Second provisional image acquisition means for
According to the specified viewpoint direction, an image obtained by interpolating each vertex coordinate of the first temporary image and each vertex coordinate of the second temporary image is specified with respect to the display object. The image processing system according to claim 7, wherein the image processing system is generated as an interpolated image of the display body when the gazing point is viewed from a viewpoint position in a specified viewpoint direction. A plurality of regions divided in a latitude direction and a meridian direction are set in a spherical virtual region centered on a predetermined center point in the display object, and the first model data is the first viewpoint direction. Data for forming the first model as a three-dimensional model of the display object when the viewpoint position is in any one of the viewpoint directions of the first region including the second model data The second as the three-dimensional model of the display object when the viewpoint position is in the viewpoint direction of any one of the second areas located in the circumferential direction around the center point. Data to form a model of
The interpolation image generation means corresponds to the interpolation viewpoint direction when the viewpoint position is in the interpolation viewpoint direction included in the interpolation area existing between the first area and the second area with respect to the circumferential direction. An interpolated image is generated by interpolating each vertex coordinate obtained based on the first model data and each vertex coordinate obtained based on the second model data, and rendered in a predetermined rendering area To
The image processing system according to claim 1.   The image processing system according to claim 1, wherein the display object is a game character appearing in a predetermined game. Based on a three-dimensional model of the display target object arranged in the virtual three-dimensional space, the computer displays an image of the display target object on the display unit according to the viewpoint direction that is the direction of the viewpoint position with respect to the display target object. An image processing method comprising:
First model data for forming a first model used as a three-dimensional model of the display object when the viewpoint position is in a first viewpoint direction with respect to the display object on the computer; When the viewpoint position with respect to the display object is in a second viewpoint direction in which the first viewpoint direction moves in a predetermined circumferential direction around the display object, the three-dimensional of the display object Storing second model data to form a second model to be used as a model;
An interpolation image that is a two-dimensional image of the display object corresponding to an interpolation viewpoint direction existing between the first viewpoint direction and the second viewpoint direction with respect to the circumferential direction is obtained from the first model data. Generated by performing an interpolation process based on each vertex coordinate of the obtained first model data and each vertex coordinate of the second model obtained from the second model data, and drawn in a predetermined drawing area And steps to
Displaying the interpolated image drawn in the drawing area on the display unit as the display object corresponding to the interpolated viewpoint direction. Based on the three-dimensional model of the display object placed at a predetermined position in the virtual three-dimensional space, the image of the display object is displayed on the display unit according to the viewpoint direction that is the direction of the viewpoint position with respect to the display object. An image processing system,
A model data storage unit that stores reference model data for forming a reference model used as a three-dimensional model of the display object when a viewpoint position with respect to the display object is in a predetermined reference direction;
As a three-dimensional model of the display object corresponding to the moving viewpoint direction when the viewpoint position with respect to the display object is in a moving viewpoint direction in which the reference direction is moved in the circumferential direction around the display object The vertex coordinates of a plurality of polygons constituting the movement model used are obtained based on the vertex coordinates of the plurality of polygons constituting the reference model and the movement amount of the moving viewpoint direction from the reference direction. A function storage unit that stores the model formation function set in accordance with the moving viewpoint direction;
Specify which moving viewpoint direction stored in the function storage unit the viewpoint position for viewing the display object is, select the model forming function corresponding to the specified moving viewpoint direction, and move the movement A movement model generating means for generating the movement model at the predetermined position by the selected model forming function based on the quantity and each vertex coordinate of the reference model;
Drawing means for drawing a movement model image obtained by viewing the generated movement model from the viewpoint position in a predetermined drawing area;
Drawing display means for displaying the movement model image drawn in the drawing area as an image of the display target object when the display target object is viewed from the viewpoint position, on the display unit; Image processing system. In the function storage unit, for each of a plurality of function sections provided in the circumferential direction of the display object, the movement based on the movement amount in the movement viewpoint direction and each vertex coordinate of the reference model The model forming function set to obtain each vertex coordinate of the model is associated,
15. The movement model generation unit generates the movement model by the selected model formation function based on each vertex coordinate of the reference model data and the movement amount. Image processing system. In the reference model data, vertex coordinates of a plurality of polygons constituting the reference model are set by being grouped into a plurality of groups,
In the function storage unit, for each group, a model forming function for obtaining each vertex coordinate of the movement model from each vertex coordinate of the reference model corresponding to the group is set. The image processing system according to claim 14 or 15. A reference axis extending in a predetermined direction through the display object is set,
The circumferential direction is a direction around the reference axis,
The movement model generation unit includes a viewpoint direction specifying unit that specifies a direction of a viewpoint position with respect to the reference axis as a viewpoint direction with respect to the display object, and the viewpoint direction specified by the viewpoint direction specifying unit is The image processing system according to claim 14, wherein the movement model is generated by specifying the movement viewpoint direction. A model data storage unit for storing reference model data for forming a reference model used as a three-dimensional model of the display object when the viewpoint position with respect to the display object is in a predetermined reference direction; and the display object Used as a three-dimensional model of the display object corresponding to the moving viewpoint direction when the viewpoint position with respect to the body is in a moving viewpoint direction in which the reference direction is moved in the circumferential direction around the display object The vertex coordinates of a plurality of polygons constituting the movement model are set so as to be obtained based on the vertex coordinates of the plurality of polygons constituting the reference model and the movement amount of the moving viewpoint direction from the reference direction. A function storage unit that stores the model forming function stored in accordance with the moving viewpoint direction, before being placed at a predetermined position in the virtual three-dimensional space. Based on the three-dimensional model of the display object, an image processing method for displaying on a display unit an image of the display object in accordance with the viewpoint direction is the direction of the viewpoint position with respect to the display object,
In the computer,
It is specified in which moving viewpoint direction stored in the function storage unit the viewpoint position for viewing the display object, the model forming function corresponding to the specified moving viewpoint direction is selected, and the movement amount And generating the movement model at the predetermined position by the selected model forming function based on each vertex coordinate of the reference model;
Drawing a movement model image obtained by viewing the generated movement model from the viewpoint position in a predetermined drawing area;
Displaying the movement model image drawn in the drawing area on the display unit as an image of the display target object when the display target object is viewed from the viewpoint position. Image processing method.

Images