JP4387363B2 - Image processing program, image processing method, and image processing apparatus - Google Patents

Description

  The present invention relates to an image processing program, an image processing apparatus, and an image processing method, and more particularly to an image processing program, image processing, and image processing method using polygons.

  Conventionally, when an afterimage is represented in a three-dimensional image, a plurality of three-dimensional models made of polygons are prepared, a plurality of three-dimensional models are rendered into a two-dimensional image for each frame, and the transparency of each three-dimensional model is set. It is changing. In the conventional afterimage representation of a three-dimensional image, the rendering image is totally translucently processed in units of polygons. However, since transparency processing is performed in units of polygons, there is a problem that the processing load is large. In addition, there is a method of expressing the transparency of the display image using the transparency (alpha value) of the texture as it is, but in this case, there is a problem that the transparency of the display image is poorly expressed.

Conventional three-dimensional image processing is described in Patent Document 1, for example.
JP 2002-298152 A

  It is an object of the present invention to improve the expression of transparency while preventing an increase in processing load in 3D image processing.

  An image processing program according to a first aspect of the present invention is a first rendering means for determining a position and an angle of a three-dimensional model and rendering at a first camera position to generate a first rendered image, and one or a plurality of polygons A model reading unit that reads a planar model configured from the storage unit, a first rendering image is pasted on the planar model, and transparency is set to at least some vertices of one or a plurality of polygons. Second rendering means for setting the transparency of each pixel on the planar model based on the transparency of the first model, assigning the transparency set to each pixel to each pixel of the first rendering image, and rendering the first rendering image A planar model is arranged according to the position and angle of the three-dimensional model corresponding to, so that it is perpendicular to the line of sight of the first camera position A planar model is rotated, and a third rendering unit for rendering the first rendering image pasted to the flat model, a.

  Here, the three-dimensional model is, for example, a three-dimensional object composed of a plurality of polygons. The three-dimensional object includes a character object and a background object. The first rendered image is an image obtained by converting the three-dimensional model into a two-dimensional image at the first camera position. The first camera position is a virtual camera position for defining a line of sight for viewing a display image in one or a plurality of frames that are currently displayed. A frame means a still image for a unit still screen. In other words, the first camera position is a camera position in a frame for displaying an image rendered by the first rendering means. When a plurality of images are rendered by the first rendering means and each image is displayed in a plurality of frames, the first camera position is a camera position common to the plurality of frames.

The planar model is composed of one or a plurality of polygons. The polygon may have any shape such as a triangular polygon or a rectangular polygon. The planar model can be constituted by, for example, a planar model having a shape of “day”, that is, two rectangular regions adjacent to each other with one side in common. The planar model is not limited to the case of two rectangular regions, and may be a shape in which a plurality of rectangular regions are arranged adjacent to each other in one direction.
The “day” -shaped planar model has two regions, and each region is, for example, a rectangle. Each region may be composed of one or a plurality of polygons. Further, the planar model can be constituted by a planar model having a shape of a “rice”, that is, a shape in which four rectangular areas are adjacent to each other with one vertex in common. The planar model of the “rice” shape has four regions, and each region is, for example, a rectangle. The planar model is not limited to the case of four rectangular regions, and may be a shape in which a plurality of rectangular regions are arranged adjacently along two directions. Each region may be composed of one or a plurality of polygons. The storage means may be any of a non-volatile storage device such as a CD-ROM and a flash memory, and a volatile storage device such as a RAM.

  The second rendering unit pastes the first rendering image on the planar model. The second rendering unit may assign texture data to each pixel of the first rendering image pasted on the planar model. The texture data is color data, for example. Texture data can be assigned to each pixel of the polygon constituting the three-dimensional model and stored in advance. Each pixel of the first rendering image has a corresponding three-dimensional model pixel. The color data of the corresponding three-dimensional model pixel is assigned to each pixel of the first rendered image.

  Further, the second rendering means sets the transparency to at least some of the vertices of one or a plurality of polygons constituting the planar model. Transparency is a scale indicating the rate at which an image transmits light from a virtual light source, and is set by an alpha value, for example. The alpha value can be set to take a value of 0 or more and 100 or less, for example. For example, when alpha value = 0, it is opaque that does not transmit light at all, and when alpha value = 100, it is transparent that transmits light completely. When the alpha value is between 0 and 100, the value is It is possible to make it translucent according to.

  The second rendering means sets the transparency of each pixel of the planar model based on the transparency set at the vertex of the polygon. The transparency of each pixel of the planar model is determined according to the distance between the pixel of the planar model and the vertex of the polygon, and the transparency set at the vertex. For example, when the first alpha value is set to the first vertex and the second alpha value is set to the second vertex, and the alpha value of the first pixel is determined, the first pixel and the second alpha value are determined. The distance d1 between the first vertex, the alpha value of the first vertex (first alpha value α1), the distance d2 between the first pixel and the second vertex, and the alpha value of the second vertex ( It is possible to determine in consideration of the second alpha value α2). For example, the alpha value of each pixel is calculated by weighted averaging the alpha values of the vertices according to the distance. If d1 + d2 = d, it can be calculated by the following equation (1).

α1 × (d1 / d) + α2 × (d2 / d) (1)
The second rendering unit assigns the transparency of each pixel of the planar model to each pixel of the corresponding first rendering image. The transparency of each pixel of the planar model is, for example, the transparency calculated by Expression (1).

  The third rendering unit arranges the planar model in accordance with the position and angle of the three-dimensional model corresponding to the first rendering image. In other words, the third rendering unit arranges the planar model in accordance with the position and angle of the three-dimensional model when the first rendering unit renders the three-dimensional model. The third rendering unit rotates the planar model so that the arranged planar model is perpendicular to the line of sight of the first camera position. Thereafter, the third rendering means renders the first rendering image attached to the planar model at the first camera position.

  The line-of-sight and the planar model are preferably vertical, but need not be completely vertical, and the first rendering image pasted on the planar model is within a range where it can be appropriately displayed. The angle between the line of sight and the planar model may deviate from the vertical. For example, if the first rendering image is within a range where a picture written on paper can be seen to be thin when viewed from the side, the angle between the line of sight and the planar model is vertical. It may be displaced. That is, the planar model can be arranged in a range substantially perpendicular to the line of sight in consideration of the rendered image.

  Note that output means for outputting a rendered image by the third rendering means can further be provided. The output unit can be, for example, a display unit that displays a rendered image by the third rendering unit. In addition, when there is an image processing apparatus that executes this program and the rendered image by the third rendering means is transferred to an external apparatus, it can be a communication unit for communicating with the external apparatus. .

  According to the first invention, the rendering data of the three-dimensional model is pasted on the planar model, and the virtual three-dimensional image is arranged so that the line of sight of the first camera position and the planar model are perpendicular to each other. Therefore, it is possible to express a virtual three-dimensional image while suppressing the processing load.

  Also, transparency is set for at least some vertices of the polygons that make up the planar model, and the transparency of each pixel of the planar model is calculated and set based on the transparency set for the vertices. The transparency of each pixel can be easily changed by changing the transparency, or by changing the coordinates of the vertexes of the polygon to deform the polygon. In addition, since the transparency of each pixel is changed according to the transparency set for the vertex and the distance from the vertex, the transparency of the image is changed in various ways compared with the case where the transparency is changed uniformly. It is possible. In other words, since the transparency of each pixel is changed according to the transparency set at the vertex and the distance from the vertex, the transparency of the image is more finely compared with the case where the transparency is uniformly changed, that is, It is possible to change with higher accuracy.

  An image processing program according to a second invention is the image processing program according to the first invention, wherein the first rendering means changes the position and angle of the three-dimensional model, so that a plurality of first Generate a rendered image. The model reading unit reads a plurality of planar models respectively corresponding to the first rendering images. The second rendering means pastes each first rendering image to each planar model, and sets transparency to at least some of the vertices of the polygon constituting the planar model corresponding to each first rendering image. The transparency of each pixel on the planar model is set based on the transparency of the vertex, and the transparency set for each pixel is assigned to each pixel of the first rendering image for rendering. The third rendering unit arranges each planar model according to the position and angle of the three-dimensional model corresponding to each first rendering image, and each planar shape is perpendicular to the line of sight of the first camera position. The model is rotated to render the first rendering data pasted on the plurality of planar models.

  The first rendering means generates a plurality of first rendering images. The second rendering unit pastes each first rendering image on a separate planar model. Further, the second rendering means sets the transparency for each vertex of the polygon for each planar model to which each first rendering image is pasted, and sets the transparency of each pixel based on the transparency of each vertex. . The second rendering means assigns the transparency set for each pixel to each pixel of the first rendering image and renders the result. The third rendering means arranges each planar model in accordance with the position and angle of the corresponding three-dimensional model, and rotates each planar model so as to be perpendicular to the camera line of sight of the current frame for rendering. .

  According to the second aspect of the invention, it is possible to display various images while suppressing an increase in processing load by pasting and rendering each first rendering image on a separate planar model. In other words, by rendering the first rendering image by pasting it on a separate planar model, the transparency can be set finely, that is, with high accuracy for each first rendering image, and an increase in processing load is suppressed. Can be displayed.

  In addition, since a plurality of first rendering images can be displayed in one frame and the transparency can be easily changed for each first rendering image, one of the plurality of first rendering images can be used as another real image. One rendering image can be expressed as an afterimage. If the first rendering image expressed as a real image among the plurality of first rendering images is sequentially changed, it is possible to express the fast movement of the object at the same frame frequency.

  An image processing program according to a third invention is the image processing program according to any one of the first and second inventions, wherein the second rendering means is configured such that the transparency of each pixel on the planar model is set to the vertex, It is set based on the distance from the vertex to each pixel.

  According to the third invention, it is possible to easily and variously change the transparency of each pixel on the planar model based on the transparency of each vertex of the polygon constituting the planar model. For example, it can be calculated by the weighted average calculation according to the above-described equation (1). In other words, the transparency of each pixel on the planar model can be changed easily and more finely, that is, with higher accuracy, based on the transparency of each vertex of the polygon constituting the planar model. For example, it can be calculated by the weighted average calculation according to the above-described equation (1).

  An image processing program according to a fourth aspect is the image processing program according to the third aspect, wherein the second rendering means changes the shape of the planar model by changing the positions of at least some vertices of the polygon.

  The planar model is deformed by changing the positions of the vertices of the polygons constituting the planar model. Thereby, even if the same transparency is set for each vertex, the distance between each pixel of the planar model and the vertex changes, so that the transparency of each pixel can be changed. As a result, it is possible to variously change the transparency of the first rendered image. In other words, the transparency of the first rendered image can be changed more finely, that is, with higher accuracy. Also, if the shape of the planar model changes, the correspondence between each pixel of the planar model and each pixel of the first rendering image can change, and this also makes the transparency of the first rendering image various. It is possible to change. In other words, if the shape of the planar model changes, the correspondence between each pixel of the planar model and each pixel of the first rendering image can change, which also increases the transparency of the first rendering image. It is possible to change it finely, that is, with higher accuracy.

  The image processing program according to the fifth invention is the image processing program according to the third invention, wherein the model reading means selects one or more planar models from a plurality of models stored in advance. Set multiple planar models.

  Similarly to the case where the planar model is deformed by selecting and reading out a plurality of pre-stored planar models and changing the planar model to which the first rendering image is pasted, the first rendering image It is possible to change the transparency in various ways. In other words, as in the case where the planar model is deformed by selecting and reading out a plurality of planar models stored in advance and changing the planar model to which the first rendering image is pasted, the first model It is possible to change the transparency of the rendered image more finely, that is, with higher accuracy.

  An image processing program according to a sixth invention is the image processing program according to any one of the first to fifth inventions, wherein the third rendering means arranges a plurality of planar models so that a plurality of rendered images are arranged continuously. By doing so, display control of the afterimage of the object is performed.

  The second rendering means pastes the plurality of first rendering images to separate planar models, and the third rendering means arranges the plurality of first rendering images so that each of the first rendering images is arranged continuously. By arranging each planar model by changing the transparency of one rendering image, it is possible to represent an afterimage when the object represented by the first rendering image moves. For example, the change in the transparency increases the transparency of the first rendered image of the previous frame as compared to the first rendered image of the subsequent frame. Here, the subsequent frame is a frame that is displayed at a relatively late time among a group of frames composed of a plurality of frames, and the previous frame is a relatively early frame among a group of frames composed of a plurality of frames. It is a frame displayed at the time. For example, in a frame group including the first to third frames, when the first frame, the second frame, and the third frame are displayed in this order, the first frame is a frame before the second frame, The second frame is a frame after the first frame, the second frame is a frame before the third frame, and the third frame is a frame after the second frame. That is, the first frame is the earliest frame and the third frame is the last frame. Alternatively, the transparency of the first rendering image of one or more frames from the last frame is decreased, and the transparency of the other first rendering image is increased. According to this, at the point where the object has already passed, it is almost transparent and can be visually recognized as an afterimage.

  An image processing program according to a seventh aspect is the image processing program according to the sixth aspect, wherein the third rendering means arranges the plurality of planar models so that one ends of the plurality of first rendering images are connected.

  If one end of the plurality of first rendering images is connected, the operation of the object composed of the plurality of joints can be expressed in a simple and diverse manner with the connection location of each first rendering image as a joint. For example, it is possible to simply express how the chains swell. In other words, if one end of a plurality of first rendering images is connected, the operation of an object composed of a plurality of joints can be expressed easily and with high accuracy using the connection location of each first rendering image as a joint. It becomes. For example, it is possible to simply express how the chains swell.

  An image processing program according to an eighth aspect of the present invention is the image processing program according to any one of the first to seventh aspects of the present invention, wherein the second rendering means displays the first rendered image rendered in any frame before the current frame, Affix to one or more planar models of the frame.

  The first rendering image already rendered in the previous frame is pasted on the planar model used for display in the current frame. In the case of afterimage representation, the same rendered image may be repeatedly displayed over a plurality of frames. In such a case, the processing load can be reduced by executing the afterimage expression by using the first rendered image rendered before again. Even when the same first rendering image is used, the transparency of each pixel of the planar model can be changed in the same manner as described above, and the transparency of the first rendering image can be changed in various and flexible ways. It is possible. In other words, even when the same first rendering image is used, the transparency of each pixel of the planar model can be changed in the same manner as described above, and the transparency of the first rendering image is made finer, ie, more It is possible to change with high accuracy and flexibility.

  An image processing program according to a ninth invention is the image processing program according to any one of the first to eighth inventions, wherein the second rendering means changes the transparency set to the vertex of the planar model for each frame.

  By changing the transparency set to the vertex of the planar model for each frame, the afterimage can be expressed more realistically. For example, afterimages can be expressed more realistically by increasing the transparency of the previous frame so that the object becomes gradually opaque. When a plurality of first rendering images are displayed in one frame, the first rendering image is displayed as a real image by gradually reducing the thickness of the other first rendering image using one of the first rendering images as a real image. Change the rendered image frame by frame. This makes it possible to display more afterimages while suppressing the processing load at the same frame frequency.

  According to a tenth aspect of the present invention, there is provided an image processing method in which a first rendering means determines a position and an angle of a three-dimensional model, renders the first three-dimensional model at a first camera position, and generates a first rendered image; Means for reading a planar planar model composed of one or a plurality of polygons; and a second rendering means for pasting the first rendering image on the planar model and for the one or a plurality of polygons Set transparency for at least some of the vertices, set the transparency of each pixel on the planar model based on the transparency of the vertices, and assign the transparency set for each pixel to each pixel of the first rendered image The rendering step, and the third rendering means is arranged in a plane in accordance with the position and angle of the three-dimensional model corresponding to the first rendered image. The model is arranged, including by rotating the planar model so that the line of sight perpendicular to the first camera position, the method comprising: rendering a first rendering image pasted to the flat model, a.

  According to the image processing method of the tenth invention, the same effects as the first invention are achieved.

  An image processing apparatus according to an eleventh aspect of the invention is characterized in that a first rendering unit that determines a position and an angle of a three-dimensional model and renders at a first camera position to generate a first rendered image, and 1 or A model reading means for reading a planar model composed of a plurality of polygons, a first rendering image is pasted on the planar model, and transparency is set to at least some vertices of one or a plurality of polygons. Second rendering means for setting the transparency of each pixel on the planar model based on the transparency of the first model, assigning the transparency set to each pixel to each pixel of the first rendering image, and rendering the first rendering image A planar model is arranged according to the position and angle of the three-dimensional model corresponding to, so that it is perpendicular to the line of sight of the first camera position The planar model is rotated, and a third rendering unit for rendering the first rendering image pasted to the flat model, a. The storage unit may be provided in the image processing apparatus or may be a storage medium that is detachable from the image processing apparatus. The storage unit may be provided in a device different from the image processing device. In this case, the model reading unit reads the planar model from the storage unit of another device via the communication line.

  The image processing apparatus according to the eleventh aspect has the same operational effects as the first aspect.

  According to the present invention, in three-dimensional image processing, it is possible to improve the expression of transparency while preventing an increase in processing load.

(1) First Embodiment FIG. 1 is a schematic configuration diagram of an image processing apparatus 1 according to an embodiment of the present invention.

  The image processing apparatus 1 includes a control unit 10, a storage unit 20, an image display unit 30, an audio output unit 40, an operation input unit 50, a communication unit 60, and a bus 70.

  The control unit 10 is configured to control the operation of the entire image processing apparatus 1. The control unit 10 transmits and receives signals to and from the storage unit 20, the image display unit 30, the audio output unit 40, the operation input unit 50, and the communication unit 60 via the bus 70, and each unit 20, 30, 40, 50, 60. To control.

  The control unit 10 can be configured by, for example, a CPU 11, a signal processor 12, and an image processor 13.

  The CPU 11 transmits and receives signals to and from the storage unit 20, the image display unit 30, the audio output unit 40, the operation input unit 50, and the communication unit 60 via the bus 70, and directly transmits each unit 20, 30, 40, 50, 60. At the same time, by controlling the signal processor 12 and the image processor 13, at least one of an image and sound is generated, and the generated image is output to the image display unit 30. The CPU 11 functions as a first rendering unit 2303 ′, a model reading unit 2304 ′, a second rendering unit 2305 ′, and a third rendering unit 2306 ′ (not shown) by executing an image processing program described later. The CPU 11 appropriately outputs a control signal to the signal processor 12 and the image processor 13 and causes the signal processor 12 and the image processor 13 to execute processing, thereby allowing the first rendering unit 2303 ′ and the model reading unit 2304 to execute the processing. ', The second rendering unit 2305', and the third rendering unit 2306 '.

  The signal processor 12 is mainly used for calculation in a three-dimensional space, calculation for conversion from a position in a three-dimensional space to a position in a pseudo three-dimensional space, light source calculation processing, and generation and processing of audio data. Process. The signal processor 12 may constitute a part of the first rendering unit 2303 ′, the model reading unit 2304 ′, the second rendering unit 2305 ′, and the third rendering unit 2306 ′.

  Based on the calculation result in the signal processor 12, the image processor 13 writes image data to be drawn in a display area of a RAM 21 (described later) of the storage unit 20, for example, an area of the RAM 21 designated by a polygon. Write texture data to The image processor 13 may constitute a part of the first rendering unit 2303 ′, the model reading unit 2304 ′, the second rendering unit 2305 ′, and the third rendering unit 2306 ′.

  The storage unit 20 functions as a storage unit that stores the image processing program, and also functions as a work area of the control unit 10 when the image processing program is executed. As will be described later with reference to FIG. 2, the image processing program includes an image processing control program 2301 and an image processing data unit 2302. Details of the image processing control program 2301 and the image processing data unit 2302 will be described later. The storage unit 20 stores calculation results and intermediate rendering images when the control unit 10 executes the image processing program. Further, the storage unit 20 stores a final rendering image calculated by the control unit 10, that is, a rendering image output to the image display unit 30.

  The storage unit 20 can be composed of, for example, a RAM 21, an interface unit 22, and a storage medium 23.

  The RAM 21 is a random access memory. The RAM 21 is, for example, a volatile storage device using a semiconductor element, and functions as a main memory (main storage device) of the control unit 10. The RAM 21 functions as a work area for the control unit 10 when executing the image processing program. The RAM 21 stores calculation results and intermediate rendered images when the control unit 10 executes the image processing program. Further, the RAM 21 stores a final rendering image calculated by the control unit 10, that is, a rendering image output to the image display unit 30. As the main storage device, a nonvolatile storage device can be used instead of a volatile storage device such as a RAM. Whether the main storage device is a volatile storage device or a non-volatile storage device, it is preferable that the main storage device has an operation speed necessary to appropriately execute the three-dimensional image processing according to the present embodiment.

  The storage medium 23 is a storage medium that is detachable from the image processing apparatus 1. The storage medium 23 may be, for example, an optically readable storage medium such as a CD-ROM or a storage medium using a semiconductor element such as a ROM or a flash memory. The storage medium 23 is preferably a non-volatile storage device in order to store an image processing program to be described later.

  The interface unit 22 interfaces transmission / reception of signals between the storage medium 23 and the bus 70. The interface unit 22 converts the signal read from the storage medium 23 into a signal receivable by the bus 70 and outputs the signal to the bus 70, and converts the signal received from the bus 70 into a signal receivable by the storage medium 23. To the storage medium 23. The control unit 10 reads the image processing program from the storage medium 23 via the interface unit 22 and stores it in the RAM 21. Actually, the CPU 11 can control data transmission / reception between the storage medium 23 and the RAM 21 by the DMAC by controlling a DMAC (Direct Memory Access Controller) or the like.

  The display control unit 30 displays a rendering image obtained as a result of image processing by the control unit 10 by executing an image processing program. The display control unit 30 receives and displays the final rendered image written in the display area of the RAM 21 that functions as a frame memory.

  The display control unit 30 can be configured by, for example, a monitor 31 and an interface unit 32.

  The monitor 31 may be composed of any flat panel such as a cathode ray tube (CRT) or a liquid crystal display panel (LCD). The monitor 31 displays a rendering image as a result of image processing by the control unit 10 by executing an image processing program. The monitor 31 receives and stores the final rendered image written in the display area of the RAM 21 that functions as a frame memory. When displaying a signal expressed as an analog signal on the monitor 31, it is preferable to provide a digital / analog conversion unit between the monitor 31 and the interface unit 32. The interface unit 32 interfaces transmission / reception of signals between the monitor 31 and the bus 70. The interface unit 32 converts the signal read from the RAM 21 into a signal that can be received by the bus 70 and outputs the signal to the bus 70, and converts the signal received from the bus 70 into a signal that can be received by the monitor 31. At the same time, the signal from the monitor 31 is subjected to predetermined conversion and output to the bus 70.

  The audio output unit 40 receives the audio data generated by the process of the control unit 10 via the bus 70, or reads the audio data generated by the process of the control unit 10 and temporarily stored in the RAM 21. After performing digital / analog conversion processing and amplification processing, it is converted into sound and output. The audio output unit 40 can include, for example, an interface unit 41, a digital / analog conversion unit 42, an amplifier 43, and a speaker 44.

  The interface unit 41 interfaces transmission / reception of signals between the digital / analog conversion unit 42 and the bus 70. The interface unit 41 converts audio data (digital signal) received via the bus 70 into a digital signal that can be received by the digital / analog conversion unit 42 and outputs the digital signal to the digital / analog conversion unit 42.

  The digital / analog conversion unit 42 converts audio data (digital signal) received via the interface unit 41 into an analog signal and outputs the analog signal to the amplifier 43.

  The amplifier 43 amplifies the audio data converted into an analog signal by the digital / analog conversion unit 42 and outputs the amplified audio data to the speaker 44.

  The speaker 44 converts the sound data amplified by the amplifier 43 into sound and outputs the sound.

  The operation input unit 50 receives an operation input from the user and outputs an operation signal corresponding to the operation to the control unit 10 via the bus 70. The operation input unit 50 can be configured by, for example, an interface unit 51 and an operation unit 52. The interface unit 51 controls transmission / reception of signals between the bus 70 and the operation unit 52. The interface unit 51 converts the operation signal output from the operation unit 52 into an operation signal that can be received by the bus 70, and outputs the operation signal to the bus 70. The operation unit 52 includes operation target units such as a push button and a lever that receive an operation from the user. The operation target unit may have any configuration such as various touch pads and voice recognition means as long as the operation from the user can be output as an operation signal.

  The communication unit 60 is configured to control communication between the image processing apparatus 1 and a communication line. The communication unit 60 outputs a signal received from the external device to the bus 70. The communication line may be either wired or wireless. For example, the communication unit 60 can receive an image processing program from a communication line and store the image processing program in the RAM 21.

  FIG. 2 is a diagram illustrating a configuration example of an image processing program.

  For example, the image processing program 2300 is stored in the storage medium 23, read from the storage unit 23 by the control unit 10, and stored in the RAM 21. The image processing program can be read from the RAM 21 by the control unit 10 and executed. The image processing program can be received via the communication unit 60 and stored in the RAM 21, and can be read from the RAM 21 and executed by the control unit 10.

  The image processing program 2300 determines the position and angle of the three-dimensional model and renders it at the first camera position to generate a first rendered image. The first camera position is a virtual camera position for defining a line of sight for viewing a display image in one or a plurality of frames that are currently displayed. A frame means a still image for a unit still screen. In other words, the first camera position is a camera position in a frame for displaying an image rendered by the first rendering means. When a plurality of images are rendered by the first rendering means and each image is displayed in a plurality of frames, the first camera position is a camera position common to the plurality of frames.

  The image processing program 2300 also reads a planar model composed of one or a plurality of polygons from the storage means (storage medium 23 or RAM 21).

  Further, the image processing program 2300 pastes the first rendering image to the planar model, sets transparency for each vertex of one or a plurality of polygons, and sets the transparency on the planar model based on the transparency of each vertex. The transparency of each pixel is set, and the transparency set for each pixel is assigned to each pixel of the first rendering image for rendering.

  The image processing program 2300 arranges the planar model in accordance with the position and angle of the three-dimensional model corresponding to the first rendering image, and the planar shape is perpendicular to the line of sight of the first camera position. The model is rotated to render the first rendering image pasted on the planar model.

  The image processing program 2300 can include an image processing control program 2301 and an image processing data unit 2302.

  The image processing control program 2301 executes the above-described image processing using data stored in the image processing data unit 2302. The image processing program 2301 includes first rendering means 2303, model reading means 2304, second rendering means 2305, and third rendering means 2306. The first rendering unit 2303 causes the control unit 10 to function as a first rendering unit. The model reading unit 2304 causes the control unit 10 to function as a model reading unit. The second rendering unit 2305 causes the control unit 10 to function as a second rendering unit. The third rendering unit 2306 causes the control unit 10 to function as a third rendering unit.

  The image processing data portion 2302 is data used when the image processing program 2301 executes image processing. The image processing data unit 2302 includes a three-dimensional model data unit 2307, a planar model data unit 2308, and a texture data unit 2309. The 3D model data unit 2307 includes a plurality of types of 3D model data. The planar model data unit 2308 includes data for designating a planar model composed of one or a plurality of polygons. The texture data unit 2309 stores texture data. The three-dimensional model data unit 2307, the planar model data unit 2308, and the texture data unit 2309 will be described in detail below.

  The 3D model data unit 2307 includes a plurality of types of 3D model data.

  FIG. 3 is an example of 3D model data included in the 3D model data unit 2307. FIGS. 3A and 3B are three-dimensional models 1001 and 1002 representing a human arm in a plurality of states. The three-dimensional model data for specifying the three-dimensional models 1001 and 1002 is composed of coordinate data for designating vertices of a plurality of polygons included in the three-dimensional models 1001 and 1002. The three-dimensional model 1001 is a three-dimensional model with arms extended, and the three-dimensional model 1002 is a three-dimensional model with arms bent.

  The planar model data unit 2308 includes data for designating a planar model composed of one or a plurality of polygons.

  4 to 6 are diagrams illustrating specific examples of the planar model included in the planar model data unit 2308.

  FIG. 4A shows a planar model 80 as a first example of the planar model.

  The planar model 80 is composed of four triangular polygons 81-84. The polygon 81 includes vertices P81, P83, and P84. The polygon 82 includes vertices P81, P82, and P84. Polygon 83 includes vertices P83, P85, and P86. Polygon 84 includes vertices P83, P84, and P86.

  The planar model 80 includes a rectangular first area composed of a polygon 81 and a polygon 82, and a rectangular second area composed of a polygon 83 and a polygon 84. The first area and the second area are partitioned by a side e82. The first region includes a side e81 as an end facing the side e82, and the second region includes a side e83 as an end facing the side e82. The planar model 80 is a planar model having a shape of “day” composed of a first region and a second region, that is, a shape in which two rectangular regions are adjacent to each other with one side in common. The planar model 80 is not limited to the case of two rectangular regions, and may be a shape in which a plurality of rectangular regions are arranged adjacently along one direction.

  The planar model data that designates the planar model 80 is composed of coordinate data that designates the coordinates of each vertex of each of the polygons 81 to 84. As will be described later, an alpha value as transparency is assigned to at least some of the vertices P81, P82, P83, P84, P85, and P86 of the polygons included in the planar model 80.

  FIG. 7 shows a specific example when alpha values are assigned to the vertices P81, P82, P83, P84, P85, and P86. In the present embodiment, the alpha value takes a value between 0 and 100. When the alpha value is 0, it is opaque that does not transmit light at all. When the alpha value is 100, it is transparent so that light is completely transmitted. When the alpha value is between 0 and 100, it depends on the value. Translucent.

  As shown in the planar model 80 in FIG. 7, in this example, alpha value = 0 is assigned to the vertices P81, P82, P83, and P84, and alpha value = 100 is assigned to the vertices P85 and P86.

  The graph shown in the figure represents the change of the alpha value set for each pixel of the planar model 80 from the side e81 toward e83. The alpha value of each pixel is calculated based on the distance from each vertex and the alpha value assigned to each vertex. In the example of FIG. 7, the alpha values assigned to the vertex P81 and the vertex P82 are the same, the alpha values assigned to the vertex P83 and the vertex P84 are the same, and the alpha values assigned to the vertex P85 and the vertex P86. Since the values are the same, the alpha value of each pixel does not change in the direction along the sides e81, e82, e83. Accordingly, the alpha value changes in the direction from the side e81 toward e83 as shown in FIG. As shown in FIG. 7B, the alpha value is alpha = 0 in the first region between the sides e81 and e82, and increases from the side e82 toward the side e83 until the alpha value = 100. The increase in the alpha value between the side e82 and the side e83 may change at a constant increase rate as shown by the straight line (1), or may change with a downwardly convex curve as shown by the curve (2). Alternatively, as shown by the curve (3), it may change with an upwardly convex curve.

  FIG. 4B is a planar model 80 'as an example of a planar model obtained by modifying the planar model 80 shown in FIG.

  In the planar model 80 ′, the coordinates of at least some of the vertices P81 ′ to P86 ′ of the polygons constituting the planar model 80 ′ correspond to the polygons of the planar model 80 in FIG. It is different from the coordinates of the vertex. Of the vertices P81 ′ to P86 ′, only one vertex may be different from the coordinates of the vertex of the planar model 80, or all the vertices may be different from the coordinates of the vertex of the planar model 80.

  In the planar model 80 and the planar model 80 ′ having at least some vertex coordinates different from each other, even if the same alpha value is set for each vertex, the distance between each pixel and the vertex in each planar model is Since they are different, it is possible to change the alpha value of each pixel. As a result, it is possible to variously change the alpha value of the first rendered image. In other words, the alpha value of the first rendered image can be changed more finely, that is, with higher accuracy. Further, as will be described later, when the first rendering image is pasted on each planar model, if the shape of the planar model changes, each pixel of the planar model and each of the first rendering image The correspondence with the pixels can change. This also makes it possible to change the alpha value of the first rendered image in various ways. In other words, the alpha value of the first rendered image can be changed more finely, that is, with higher accuracy. Therefore, the planar model 80 is transformed into the planar model 80 ′, and the planar model is deformed by changing the coordinates of the vertices of the polygons constituting the planar model, thereby being pasted on the planar model. It is possible to variously change the alpha value of the first rendered image. In other words, the planar model 80 is transformed into the planar model 80 ', and the coordinates of each vertex of the polygons constituting the planar model are changed to deform the planar model, thereby being pasted on the planar model. It is possible to change the alpha value of the first rendered image to be finer, that is, with higher accuracy. As a result, it is possible to variously express the transparency of the image. In other words, the transparency of the image can be expressed more finely, that is, with higher accuracy.

  FIG. 5A shows a planar model 90 as a second example of the planar model.

  The planar model 90 is composed of two rectangular polygons 91 and 92. The polygon 91 includes vertices P91, P92, P93, and P94. Polygon 92 includes vertices P93, P94, P95, and P96. The planar model 90 is a “day” -shaped polygon composed of polygons 91 and 92.

  The planar model data that designates the planar model 90 is composed of coordinate data that designates the coordinates of the vertices P91, P92, P93, P94, P95, and P96 of the polygons 91 and 92, respectively. An alpha value as transparency is assigned to at least a part of the vertices P91, P92, P93, P94, P95, and P96 of the polygon included in the planar model 90.

  FIG. 5B shows a planar model 110 as a third example of the planar model.

  The planar model is composed of one rectangular polygon 111. The polygon 111 includes vertices P111, P112, P113, and P114.

  The planar model data designating the planar model 110 is composed of coordinate data designating the vertices P111, P112, P113, and P114 of each polygon 111. Alpha values as transparency are assigned to the vertices P111, P112, P113, and P114 of the polygons included in the planar model 110.

  FIG. 6A shows a planar model 120 as a fourth example of the planar model.

  This planar model 120 is a planar model having a shape of a “field” composed of four rectangular polygons, that is, a shape in which four rectangular areas are adjacent to each other with one vertex in common. The four rectangular polygons are a polygon 121, a polygon 122, a polygon 123, and a polygon 124. Each of the polygons 121 to 124 may be composed of one or a plurality of polygons. The planar model data for specifying the planar model 120 is constituted by coordinate data for designating the coordinates of the vertices of the polygons 121 and 122. The planar model 80 is not limited to the case of four rectangular regions, and may be a shape in which a plurality of rectangular regions are arranged adjacently along two directions.

  The polygon 121 includes vertices P120, P121, P123, and P124. The polygon 122 includes vertices P121, P122, P124, and P125. The polygon 123 includes vertices 123, P124, P126, and P127. The polygon 124 includes vertices P124, P125, P127, and P128. An alpha value is set for each of the vertices P121 to P128. For example, the alpha value of each pixel included in the polygon 121 is calculated using the alpha value set for each vertex P121, P122, P124, and P125 of the polygon 121.

  The alpha value of each pixel of the planar model 120 is set based on the alpha values of the nine vertices of the vertices P120 to P128. On the other hand, the alpha value of each pixel of the planar model 80 described above is set based on the alpha values of the six vertices P81 to P86. Therefore, since the alpha value of each pixel of the planar model 120 is set based on the alpha values of more vertices than in the planar model 80, the planar model 120 has more various alpha values. Can be changed. In other words, since the alpha value of each pixel of the planar model 120 is set based on the alpha values of more vertices than in the planar model 80, the planar model 120 has a higher alpha value. More specifically, it can be changed with higher accuracy. Further, since the vertex P124 of the planar model 120 is not in the end portion of the planar model 120 but inside, it is possible to set the alpha value of each pixel in more detail than when there is a vertex only at the end portion. It is. As a result, it is possible to variously express the transparency of the image. In other words, the transparency of the image can be expressed more finely, that is, with higher accuracy.

  FIG. 6B shows a planar model 120 ′ that is a planar model 120.

  In the planar model 120 ′, the coordinates of at least some of the vertices P121 ′ to P128 ′ of the polygon constituting the planar model 120 ′ are different from the coordinates of the vertices of the polygon of the planar model 120. Of the vertices P121 ′ to P128 ′, only one vertex may be different from the coordinates of the vertex of the planar model 120, or all the vertices may be different from the coordinates of the vertex of the planar model 120.

  In the planar model 120 and the planar model 120 ′ in which the coordinates of at least some of the vertices are different, even if the same alpha value is set for each vertex, the distance between each pixel and the vertex in each planar model is Since they are different, it is possible to change the alpha value of each pixel. As a result, it is possible to variously change the alpha value of the first rendered image. In other words, the alpha value of the first rendered image can be changed more finely, that is, with higher accuracy. Further, as will be described later, when the first rendering image is pasted on each planar model, if the shape of the planar model changes, each pixel of the planar model and each of the first rendering image The correspondence with the pixels can change. This also makes it possible to change the alpha value of the first rendered image in various ways. In other words, the alpha value of the first rendered image can be changed more finely, that is, with higher accuracy. Therefore, the planar model 120 can be pasted to the planar model by transforming the planar model by changing the coordinates of each vertex of the polygon constituting the planar model 120 so as to transform the planar model 120 into the planar model 120 ′. It is possible to variously change the alpha value of the first rendered image. In other words, as the planar model 120 is transformed into the planar model 120 ', the coordinates of the vertices of the polygons constituting the planar model are changed and the planar model is deformed to be pasted on the planar model. It is possible to change the alpha value of the first rendered image to be finer, that is, with higher accuracy. As a result, it is possible to change the transparency of the image in various ways. In other words, the transparency of the image can be changed more finely, that is, with higher accuracy.

  In FIG. 2, the texture data unit 2309 stores texture data. The texture data is, for example, color data. The color data is assigned to each pixel of the polygon constituting the 3D model data. The color data is stored in association with the coordinates of each pixel of the polygon constituting the three-dimensional model data.

  The data included in the above-described three-dimensional model data unit 2307, planar model data unit 2308, and texture data unit 2309 may be read out while being stored in the storage medium 23, or once read out into the RAM 21. After being stored, it may be read from the RAM 21.

  Next, each configuration of the image processing control program 2301 will be described.

  The first rendering unit 2303 determines the position and angle of the three-dimensional model and renders it at the first camera position to generate a first rendered image.

  The first rendering unit 2303 renders the three-dimensional models 1001 and 1002 as shown in FIG. 3 at the first camera position. A rendering image obtained as a result of this rendering is defined as a first rendering image.

  For example, the first rendering unit 2303 reads the 3D model data specifying the 3D model 1002 from the 3D model data unit 2307 so that the palm faces the current camera position as shown in FIG. The first rendered image 1002a corresponding to the three-dimensional model 1002 is output after rendering at the current camera position. For example, the first rendering unit 2303 stores the first rendered image 1002a in the RAM 21. The first rendering unit 2303 reads out the three-dimensional model data specifying the three-dimensional model 1001 from the three-dimensional model data unit 2307 and arranges the palm so as to face the current camera position as shown in FIG. 8B. Then, rendering is performed at the current camera position, and a first rendered image 1001b corresponding to the three-dimensional model 1002 is output. For example, the first rendering unit 2303 stores the first rendering image 1001b in the RAM 21. The first rendering unit 2303 reads the 3D model data for specifying the 3D model 1002 from the 3D model data unit 2307 and arranges the back of the hand to face the current camera position as shown in FIG. 8C. Then, rendering is performed at the current camera position, and a first rendered image 1002c corresponding to the three-dimensional model 1002 is output. For example, the first rendering unit 2303 stores the first rendered image 1002c in the RAM 21.

  The model reading unit 2304 reads planar model data composed of one or a plurality of polygons from the planar model data unit 2308. The model reading unit 2304 can read planar model data from the storage medium 23. The model reading unit 2304 may read planar model data read and stored in the RAM 21. The planar model data is data for specifying a planar model such as the planar models 80, 90, 110, and 120 as described above. Planar model data obtained by deforming the planar model 80 (for example, the planar model 80 ′ and the planar model 120 ′) can also be stored in the planar model data unit 2308 in advance. For example, the planar model data unit 2308 can include a plurality of planar models whose coordinates of the vertices of the planar model 80 are changed.

  The second rendering unit 2305 pastes the first rendering image on the planar model. Specifically, the second rendering unit 2305 assigns each pixel of the first rendered image to each pixel of the planar model.

  For example, the second rendering unit 2305 assigns a first rendered image 1002 a obtained by rendering the three-dimensional model 1002 to each pixel of the planar model 80. The second rendering unit 2305 assigns texture data to each pixel of the planar model 80 to which each pixel of the first rendering image 1002a is assigned.

  The second rendering unit 2305 sets an alpha value for each vertex of one or a plurality of polygons, and sets an alpha value for each pixel on the planar model based on the alpha value of the vertex. For example, as shown in FIG. 7, the second rendering unit 2305 sets alpha values for the vertices P81 to P86 of the polygons 81 to 84 constituting the planar model 80. The second rendering means 2305 calculates and sets the alpha value of each pixel based on the alpha value of each vertex P81 to P86 and the distance between each pixel and each vertex P81 to P86. In the example of FIG. 7, each pixel included in the polygon 81 is calculated based on the distance from each pixel to the vertices P81 to P84 and the alpha value of each of the vertices P81 to P84. When the distance from a pixel to each vertex P81 to P84 is d81 to d84 and the sum of d81 to d84 is d, and the alpha values of the vertices P81 to P84 are α1 to α4, the alpha value of the pixel is expressed by the equation (2). ) To calculate.

α1 × (d1 / d) + α2 × (d2 / d) + α3 × (d3 / d) + α4 × (d4 / d)
... (2)
Further, the alpha value of each pixel included in the polygon 82 is calculated in the same manner.

  The second rendering unit 2305 assigns an alpha value calculated by, for example, Expression (2) and set to each pixel to each pixel of the first rendering image.

  As a result, the second rendering means 2305 assigns each pixel of the first rendered image 1001a to the planar model 80, and texture data and an alpha value are assigned to these pixels. The second rendering unit 2305 causes the RAM 21 to store the coordinates of each vertex of the polygon constituting the planar model 80 and the texture data and the alpha value associated with the coordinates of each pixel of the planar model 80.

  The third rendering unit 2306 arranges the planar model in accordance with the position and angle of the three-dimensional model corresponding to the first rendered image, and rotates the planar model so as to be perpendicular to the line of sight of the current camera position. Then, the first rendering image pasted on the planar model is rendered.

  The third rendering unit 2306 is a plane on which the first rendering image 1002a is rendered in accordance with the position and angle of the three-dimensional model 1002 (the position and angle at which the palm side faces the camera) when rendered by the first rendering unit 2303. A model 80 is placed. The third rendering unit 2306 renders the first rendered image 1001b in accordance with the position and angle of the three-dimensional model 1001 (the position and angle at which the palm side faces the camera) when rendered by the first rendering unit 2303. The planar model 80 is arranged. The third rendering unit 2306 renders the first rendered image 1002c in accordance with the position and angle of the three-dimensional model 1001 (the position and angle at which the back of the hand faces the camera) when rendered by the first rendering unit 2303. The planar model 80 is arranged.

  The third rendering unit 2306 stores a rendering image obtained as a result of the processing in the RAM 21.

  FIG. 10 is a flowchart of image processing according to the present embodiment.

  In step S101, the first rendering means 2303 sets a first camera position (hereinafter referred to as the current camera position). The first camera position is a virtual camera position for defining a line of sight for viewing a display image in one or a plurality of frames that are currently displayed. A frame means a still image for a unit still screen. In other words, the first camera position is a camera position in a frame for displaying an image rendered by the first rendering means. When a plurality of images are rendered by the first rendering means and each image is displayed in a plurality of frames, the first camera position is a camera position common to the plurality of frames.

  In step S102, the first rendering unit 2303 renders the three-dimensional models 1001 and 1002 shown in FIGS. 3A and 3B at the first camera position set in step S101.

  The first rendering unit 2303 arranges the three-dimensional model 1002 so that the palm side faces the current camera position (corresponding to FIG. 8A), and renders the three-dimensional model 1002, thereby rendering the first A rendering image 1002a is generated.

  Further, the first rendering unit 2303 arranges the three-dimensional model 1001 so that the palm side faces the current camera position (corresponding to FIG. 8B), and renders the three-dimensional model 1001, thereby rendering the first One rendering image 1001b is generated.

  The first rendering unit 2303 arranges the three-dimensional model 1001 so that the back side of the hand faces the current camera position (corresponding to FIG. 8C), and renders the three-dimensional model 1002, thereby rendering the first One rendering image 1002c is generated.

  In step S103, the model reading unit 2304 reads the planar model from the three-dimensional model data unit 2307.

  Here, the model reading unit 2304 reads the planar model 80 for pasting each of the first rendering images 1002a, 1001b, and 1002c. The model reading unit 2304 may read a planar model having a different type or the same type and a different shape for each of the first rendering images 1002a, 1001b, and 1002c. Here, the model reading unit 2304 reads the same planar model 80 for each of the first rendering images 1002a, 1001b, and 1002c. In addition, the model reading unit 2304 sets the planar model 80 read for the first rendering image 1002a as 801, and sets the planar model 80 read out for the first rendering image 1001b as 802, so that the first rendering is performed. The planar model 80 read out for the image 1002c is assumed to be 803.

  In step S104, the second rendering unit 2305 pastes the first rendering image on the planar model.

  As shown in FIG. 8A, the second rendering unit 2305 pastes the first rendering image 1002a on the planar model 801. The second rendering unit 2305 assigns each pixel of the first rendered image 1002a to each pixel of the planar model 801. The second rendering unit 2305 reads the color data of each pixel of the first rendering image 1002a from the texture data unit 2309 and assigns each pixel of the first rendering image 1002a among the pixels of the planar model 801. Color data is assigned to the selected pixel.

  As shown in FIG. 8B, the second rendering unit 2305 pastes the first rendering image 1001b on the planar model 802. The second rendering unit 2305 assigns each pixel of the first rendered image 1001b to each pixel of the planar model 802. The second rendering unit 2305 reads the color data of each pixel of the first rendering image 1001b from the texture data unit 2309, and assigns each pixel of the first rendering image 1002b among the pixels of the planar model 802. Color data is assigned to the selected pixel.

  As shown in FIG. 8C, the second rendering unit 2305 pastes the first rendering image 1002c on the planar model 803. The second rendering unit 2305 assigns each pixel of the first rendered image 1002c to each pixel of the planar model 803. The second rendering unit 2305 reads the color data of each pixel of the first rendering image 1002c from the texture data unit 2309 and assigns each pixel of the first rendering image 1002c among the pixels of the planar model 803. Color data is assigned to the selected pixel.

  8A to 8C, an object 820 indicates a shoulder joint 820. In this example, the first rendered images 1002a, 1001b, and 1002c represent movements that pivot about the shoulder joint 820 side end.

  In step S105, the second rendering unit 2305 sets alpha values for at least some of the vertices of the polygon constituting the planar model, and sets the alpha value of each pixel of the planar model based on the alpha value of the vertices. .

  For example, the second rendering unit 2305 sets an alpha value at each vertex of each polygon (polygons 81 to 84 shown in FIG. 4A) constituting each planar model 801, 802, 803. Further, the second rendering means 2305 converts the alpha value of each pixel on each planar model 801, 802, 803 into the distance from each pixel to the vertex and the alpha value set at the vertex as described above. Calculate based on

  The second rendering means 2305 sets an alpha value at the vertex of each polygon constituting the planar model 801, and the alpha value of each pixel based on the alpha value set at the vertex and the distance from the vertex to each pixel. Is calculated and set.

  The second rendering unit 2305 also sets an alpha value at the vertex of each polygon constituting the planar model 802, and based on the alpha value set at the vertex and the distance from the vertex to each pixel, Calculate and set the alpha value.

  Also, the second rendering means 2305 sets an alpha value at the vertex of each polygon constituting the planar model 803, and based on the alpha value set at the vertex and the distance from the vertex to each pixel, Calculate and set the alpha value.

  Here, assuming that the first rendering images are moved in the order of the planar models 801 to 803, the alpha value is smaller (opaque) as the first rendering image at the movement destination, that is, the first rendering image in the subsequent frame. The alpha value is set to be larger (closer to transparency) as the first rendering image of the movement source, that is, the first rendering image of the previous frame. In the first to third frames, the first rendering image 1002c of the planar model 803 is a real image, and the alpha value of the planar model 801> the alpha value of the planar model 802> the alpha value of the planar model 803.

(Generation of first frame image)
In step S106, the third rendering unit 2306 arranges the planar model in accordance with the position and angle of the three-dimensional model corresponding to the first rendering image. The third rendering unit 2306 arranges the planar model 801 on which the first rendering image 1002a is pasted in accordance with the position and angle of the three-dimensional model 1002 rendered in step S102. Next, the third rendering unit 2306 rotates the planar model 801 so as to be perpendicular to the line of sight of the current camera position. The planar model 801 arranged in this way is shown in FIG.

  In step S107, the third rendering unit 2306 rotates the planar model so as to be perpendicular to the line of sight of the first camera position, and renders the first rendered image pasted on the planar model. The third rendering unit 2306 renders the first rendering image 1002a pasted on the planar model 801. The third rendering unit 2306 stores the final image obtained by rendering the first rendering image 1002a in the RAM 21 as the first frame image.

(Generation of second frame image)
Returning to step S106 again, the second frame image is generated.

  In step S106, the third rendering unit 2306 arranges the planar model 802 on which the first rendering image 1001b is pasted in accordance with the position and angle of the three-dimensional model 1002 rendered in step S102. Next, the third rendering unit 2306 rotates the planar model 802 so as to be perpendicular to the line of sight of the current camera position. A planar model 802 arranged in this way is shown in FIG.

  In step S107, the third rendering unit 2306 renders the first rendered image 1002b pasted on the planar model 802. The third rendering unit 2306 stores the final image obtained by rendering the first rendering image 1002b in the RAM 21 as the second frame image.

(Generation of third frame image)
Returning to step S106 again, an image of the third frame is generated.

  In step S106, the third rendering unit 2306 arranges the planar model 803 on which the first rendering image 1002c is pasted in accordance with the position and angle of the three-dimensional model 1001 rendered in step S102. Next, the third rendering unit 2306 rotates the planar model 803 so as to be perpendicular to the line of sight of the current camera position. A planar model 803 arranged in this way is shown in FIG.

  In step S107, the third rendering unit 2306 renders the first rendering image 1002c pasted on the planar model 803. Then, the third rendering unit 2306 stores the final image obtained by rendering the first rendering image 1002c in the RAM 21 as the third frame image.

  As described above, the first rendered images 1002a, 1001b, and 1002c rendered at the first camera position set in step S101 are displayed in the first to third frames, respectively. Accordingly, the camera position in the first to third frames is the first camera position.

  The first to third frame images stored in the RAM 21 are output and displayed on the image display unit 30 at a predetermined cycle.

  FIG. 9 shows a state of a screen that is visually recognized when images of the first frame to the third frame are displayed on the image display unit 30. In this way, when the first frame to the third frame are continuously displayed, an animation can be expressed by the image of each frame. In this case, the alpha value of the planar model set in step S105 is set to be larger for the previous frame, that is, the alpha value of the planar model 801> the alpha value of the planar model 802> the alpha value of the planar model 803. Can be expressed by the images from the first frame to the third frame while the arm moves while leaving an afterimage. In other words, according to the three-dimensional image processing according to the present embodiment, by changing various alpha values through simple processing, it is possible to render realistic afterimages through simple processing. In other words, by changing the alpha value more finely, that is, with higher accuracy by simple processing, as a result, realistic afterimage can be expressed by simple processing.

  As described in step S105, in the present embodiment, the alpha value assigned to the texture data is not used as it is, and the alpha value of the model to which the texture is pasted is not changed as a whole. Alpha values are set for a plurality of vertices included in the model, and the alpha value of each pixel of the planar model is calculated using the alpha values of the vertices. Therefore, various alpha values can be changed with a simple calculation. In other words, the alpha value can be changed more finely, that is, with higher accuracy by simple calculation. As a result, it is possible to variously express the transparency of the image. In other words, the transparency of the image can be expressed more finely, that is, with higher accuracy. Further, if the texture pasted on the planar model is repeatedly used, the alpha value can be changed in various ways while suppressing the processing load, and as a result, a realistic afterimage can be expressed. In other words, if the texture pasted on the planar model is used repeatedly, the alpha value can be changed more finely, that is, with higher accuracy while suppressing the processing load, resulting in a realistic afterimage expression. It becomes possible.

<Display after the fourth frame>
Thereafter, when displaying the image after the fourth frame, for example, the planar model 802 is displayed in the fourth frame, the planar model 803 is displayed in the fifth frame, and the planar model 801 is displayed in the sixth frame. That is, an image is displayed by repeatedly using the planar models 801 to 803. The motion of the three-dimensional model is expressed by repeatedly using the same image in units of three frames. The first camera position set in the fourth to step S101 is used.

  In this case, the planar models “801 (afterimage), 802 (afterimage), 803 (real image)”, “801 (afterimage), 803 (afterimage), 802 (real image)”, “803 (afterimage), 802 (afterimage) , 801 (real image) ”,..., And shift the planar model of the last frame (model with the real image pasted) one by one every three frames, and after every three frames Reduce the alpha value of the planar model of the frame (make it more opaque). Specifically, in the first to third frames, the planar model 803 is the latest model and has the smallest alpha value (closest to the opacity), and in the fourth to sixth frames, the planar model 802 is the latest model. Model with the smallest alpha value (closest to opaque), and in the seventh to ninth frames, the planar model 801 is the model of the last frame and has the smallest alpha value (closest to opaque). To be.

  When displaying the fourth to sixth frames, the first rendering images 1002a, 1001b, and 1002c attached to the planar models 801 to 803 in the processing of the first to third frames are used as they are.

  However, in the fourth to sixth frames, since the planar model 802 is a planar model of the last frame and is a real image, the alpha value of the planar model 801> the alpha value of the planar model 803> the planar model. In step S105, the alpha values of the planar models 801 to 803 are reset so that the alpha value becomes 802.

  Thereafter, similarly to the case of the first to third frames, steps S106 and S107 are executed to display the fourth to sixth frames.

  When displaying the seventh to ninth frames, the first rendering images 1002a, 1001b, and 1002c attached to the planar models 801 to 803 in the processes of the fourth to sixth frames are used as they are. The camera positions of the seventh to ninth frames are the first camera positions set in step S101.

  However, in the seventh to ninth frames, since the planar model 801 is the planar model of the last frame and is a real image, the alpha value of the planar model 803> the alpha value of the planar model 802> the planar model 801. In step S105, the alpha values of the planar models 801 to 803 are reset so as to obtain the alpha value. Thereafter, similarly to the case of the first to third frames, steps S106 and S107 are executed to display the seventh to ninth frames.

  As described above, when the first to ninth frames are displayed, the real image of the three-dimensional model is the first rendered image 1002c (planar model 803), the first rendered image 1001b (planar model 802), the first The rendering image 1002a (planar model 801) can be expressed in a changing manner with an afterimage.

  Further, if the planar model 802 is displayed as a real image in the tenth to twelfth frames and the planar model 803 is displayed as a real image in the thirteenth to fifteenth frames, the arm is swung down (first to third frames), It is possible to express a series of motions of swinging up (fourth to sixth frames, seventh to ninth frames) and swinging down the arms (tenth to twelfth frames) together with afterimages. That is, if a planar model as a real image is repeatedly displayed in the order of 801, 802, 803, 802, 801, it is possible to express the motion of swinging the arm up and down together with the afterimage. The camera positions of the tenth to twelfth frames are the first camera positions set in step S101.

  The second rendering unit 2305 pastes the first rendered image rendered in any frame before the current frame on the planar model of the current frame. Alternatively, a planar model to which the first rendering image of the previous frame is pasted can be used as it is.

  For example, when displaying the first rendering image 1001b in the fourth frame, the first rendering image 1002b obtained by rendering the three-dimensional model 1002 in the second frame is pasted on the planar model 802 as it is. Alternatively, the planar model 802 to which the first rendering image 1002b is attached is used as it is. However, even in this case, the alpha value is calculated. Therefore, it is possible to perform realistic afterimage expression while reducing the processing load by reusing the image rendered in the previous frame.

  Further, by changing the alpha value set for each planar model every three frames, the alpha values of the first rendered images 1002a, 1001b, and 1002c can be flexibly changed. As a result, the alpha value can be changed in various ways while suppressing the processing load, and a realistic afterimage can be expressed. In other words, the alpha value can be changed more finely, that is, with higher accuracy while suppressing the processing load, and a realistic afterimage can be expressed.

  In the unit of three frames, the alpha value of the first rendered image of the last frame displayed as a real image is the smallest (close to opaque), and the alpha value of the previous frame is made large (close to transparency). In other words, the alpha value of the previous frame is increased (close to transparency) with respect to the frame in which the real image is displayed. This makes it possible to make the afterimage expression realistic.

<Change of alpha value by deformation of planar model>
As a method of deforming the planar model, for example, a plurality of planar models obtained by deforming the planar model 80 are stored in the storage unit 23 in advance. Then, when the planar model is read out in step S103 of FIG. 10, the planar model 80 having a different shape is appropriately read out. The deformed planar model is a planar model having a different shape, the shape of which is changed by changing the coordinates of the vertices of the polygons constituting the reference planar model. Then, by executing steps S <b> 104 and S <b> 105 for the planar model 80 having a different shape, the alpha value set for the first rendered image is compared with the case where the alpha value is set for the planar model 80 before deformation. It is possible to change.

  As another method of deforming the planar model, a method of deforming the planar model by adding step S1041 in which the second rendering unit 2305 deforms the planar model 80 between step S103 and step S104. There is. In this case, in step S1041, the planar model 80 is deformed by changing the coordinates of the vertices of the polygons constituting the planar model read in step S103. By executing steps S104 and S105 on the deformed planar model 80, the alpha value set in the first rendered image is changed as compared with the case where an alpha value is set in the planar model 80 before deformation. Is possible.

<When rendering multiple planar models in one frame>
In the above description, the third rendering unit 2306 generates one of the final images of the planar models 801, 802, and 803 for each frame. On the other hand, the third rendering unit 2306 may generate a final image of two or more planar models in one frame.

  For example, the third rendering unit 2306 may render the three planar models 801, 802, and 803 in one frame and display the three planar models in one frame. This processing is as follows, for example.

(Generation of first frame image)
In the flowchart of FIG. 10, after steps S101 to S103 are processed in the same manner as described above, the process proceeds to step S104.

  In step S104, as shown in FIG. 8A, the second rendering unit 2305 pastes the first rendered images 1002a, 1001b, and 1002c on the planar models 801, 802, and 803, respectively. The second rendering unit 2305 assigns the pixels of the first rendered images 1002a, 1001b, and 1002c to the pixels of the planar models 801, 802, and 803, respectively. The second rendering unit 2305 reads the color data of each pixel of the first rendered images 1002a, 1001b, and 1002c from the texture data unit 2309, and among the pixels of the planar models 801, 802, and 803, Color data is assigned to the pixel to which each pixel of the rendered images 1002a, 1001b, and 1002c is assigned.

  In step S105, the second rendering unit 2305 sets alpha values at least at some vertices of the polygons (polygons 81 to 84 shown in FIG. 4A) constituting the planar models 801, 802, and 803. . Further, the second rendering means 2305 converts the alpha value of each pixel on each planar model 801, 802, 803 to the distance from each pixel to at least some of the vertices and the alpha value set for the vertices. Calculate based on

  Here, assuming that the first rendering image is moved in the order of the planar models 801 to 803, the alpha value is smaller as the first rendering image to be moved, that is, the first rendering image in the later frame, and the movement is performed. The alpha value is set to be larger as the original first rendered image, that is, the first rendered image of the previous frame. Specifically, the alpha value of the planar model 801> the alpha value of the planar model 802> the alpha value of the planar model 803.

  In step S106, as shown in FIG. 9, the third rendering means 2306 converts the planar models 801, 802, 803 according to the positions and angles of the three-dimensional models corresponding to the first rendered images 1002a, 1002b, 1001c. Deploy.

  In step S107, the third rendering unit 2306 rotates the planar models 801, 802, 803 so as to be perpendicular to the line of sight of the current camera position, and pastes the planar models 801, 802, 803 on the planar models 801, 802, 803. One rendering image 1002a, 1001b, 1002c is rendered. The third rendering unit 2306 stores the final image obtained by rendering the first rendering images 1002a, 1001b, and 1002c in the RAM 21 as the first frame image.

(Generation of second frame image)
Returning to step S104, the first rendering images 1002a, 1001b, and 1002c used in the processing of the first frame are pasted on the planar models 801, 802, and 803. The camera position of the second frame is the first camera position set in step S101.

  In step S105, in each of the planar models 801, 802, and 803, an alpha value is set at the vertex of the polygon constituting the planar model, and based on the alpha value of the vertex and the distance from the vertex to each pixel. The alpha value of each pixel on the planar model is calculated and set.

  In the second frame, the planar model of the last frame, that is, the planar model of the real image is the planar model 802, and the alpha value of the planar model 801> the alpha value of the planar model 803> the alpha value of the planar model 802 And

  Thereafter, in the same manner as described above in steps S106 and S107, the rendering images of the first rendering images 1002a, 1001b, and 1002c attached to the planar models 801 to 803 are stored in the RAM 21 as the second frame.

(Generation of third frame image)
Returning to step S104, the first rendering images 1002a, 1001b, and 1002c used in the processing of the first frame are pasted on the planar models 801, 802, and 803. The camera position of the third frame is the first camera position set in step S101.

  In step S105, in each of the planar models 801, 802, and 803, an alpha value is set at the vertex of the polygon constituting the planar model, and based on the alpha value of the vertex and the distance from the vertex to each pixel. The alpha value of each pixel on the planar model is calculated and set.

  In the third frame, the planar model of the last frame, that is, the planar model of the real image is the planar model 801, and the alpha value of the planar model 803> the alpha value of the planar model 802> the alpha value of the planar model 801 And

  Thereafter, in the same manner as described above in steps S106 and S107, the rendering images of the first rendering images 1002a, 1001b, and 1002c attached to the planar models 801 to 803 are stored in the RAM 21 as the second frame.

  The first to third frame images stored in the RAM 21 are output and displayed on the image display unit 30 at a predetermined cycle. At this time, the real image changes to the planar model 803, the planar model 802, and the planar model 801 in the order of the first frame, the second frame, and the third frame, so that the real image changes with an afterimage. This can be expressed while suppressing the processing load.

  When the planar models 801, 802, and 803 are displayed every frame, the real image moves every three frames, but the three planar models are changed while changing the alpha value of the planar models 801, 802, and 803. In the case of expressing with one frame, the real image moves every frame. Therefore, it is possible to increase the moving speed of the real image in the same frame period. In particular, as the number of afterimages increases, the processing load associated with afterimage expression increases, but it is possible to suppress the processing load by displaying a plurality of planar models while changing the alpha value in this way. is there.

  In the example shown in FIGS. 8 and 9, display processing is performed so that one end of the plurality of first rendering images, that is, the base portion of the arm is connected, but one end of the plurality of first rendering images is not necessarily connected. Absent.

  FIG. 11 shows a case where an object in the shape of an arm expresses a state of flying in a direction indicated by an arrow A in the figure, and a plurality of first rendering images are continuously arranged without being connected. It is a figure explaining the process performed.

  In the figure, first rendering images 1003, 1004, and 1005 are pasted on the planar models 804, 805, and 806, respectively. Further, as indicated by an arrow B in the figure, the alpha value of the planar model of the previous frame is set to be larger. That is, the alpha value of the planar model 804> the alpha value of the planar model 805> the alpha value of the planar model 806 is set. Further, the first rendering image 1003 of the planar model 804, the first rendering image 1004 of the planar model 805, and the first rendering image 1005 of the planar model 806 are one frame by the processing of steps S101 to S107 in FIG. Display for each. As a result, afterimage representation when an object having an arm shape flies from the first rendering image 1003 toward the first rendering image 1005 can be performed by the same simple process as described above.

  FIG. 12 is another example in which one end of a plurality of first rendering images is connected. FIG. 12 is a diagram illustrating a process for expressing a state in which a chain in which a plurality of rings are connected undulates.

  FIG. 4A shows an image of the first frame, and FIG. 4B shows an image of the second frame. In FIG. 9A, first rendering images 1101a, 1102a,... Representing a ring are pasted on planar models 901a, 902a,. In FIG. 6B, first rendering images 1101b, 1102b,... Representing a ring are pasted on the planar models 901b, 902b,.

  The first rendering images 1101a, 1102a,... Are obtained by rendering the three-dimensional model corresponding to the plurality of rings in FIG. Further, the first rendered images 1101a, 1102a,... Are pasted on the planar models 901a, 902a,... And the alpha values of the respective pixels of the planar models 901a, 902a,. , By assigning the alpha value of each pixel to each pixel of the first rendered image 1101a, 1102a,..., The first rendered image 1101a, 1102a,. Render to Then, the planar models 901a, 902a,... Are arranged at the positions and angles when the three-dimensional models corresponding to the first rendered images 1101a, 1102a,. .., And then rendering the first rendered images 1101a, 1102b,... To render the final rendered image of the first frame. Get as an image.

  The first rendering images 1101b, 1102b,... Are obtained by rendering the three-dimensional model corresponding to the plurality of rings in FIG. Further, the first rendered images 1101b, 1102b,... Are pasted on the planar models 901b, 902b,... And the alpha values of the respective pixels of the planar models 901b, 902b,. , By assigning the alpha value of each pixel to each pixel of the first rendered image 1101b, 1102b,..., The first rendered image 1101b, 1102b,. Render to Here, the alpha value of each pixel of the planar models 901b, 902b,... Is made smaller than the alpha value of each pixel of the planar models 901a, 902a,. Then, the planar models 901b, 902b,... Are arranged at the positions and angles when the three-dimensional models corresponding to the first rendered images 1101b, 1102b,. After the planar models 901b, 902b,... Are rotated so as to be perpendicular to the line of sight, the first rendered images 1101b, 1102b,. Get as an image.

  When the first frame and the second frame are continuously displayed, the alpha value of the second frame is smaller than the alpha value of the first frame, so that the image of the second frame is viewed as a real image and the image of the first frame is viewed as an afterimage. It is possible to control the display so that

  Also, in the frames after the third frame, while repeatedly using the rendered images of the first frame and the second frame, the magnitude relationship of the alpha values of the images in FIGS. If the images of FIGS. 12A and 12B are displayed, it is possible to express a state in which the chain undulates while the real image and the afterimage are alternately switched in FIGS. 12A and 12B. .

  Here, the image of the two patterns of FIG. 12A and FIG. 12B is used, but the image of each pattern is repeated for every frame of the same number as the number of patterns using the images of three or more patterns. By displaying, it is possible to more realistically express the undulation of the chain.

1 is a schematic configuration diagram of an image processing apparatus according to an embodiment of the present invention. 2 is a configuration example of an image processing program. A specific example of a three-dimensional model. A specific example of a planar model. A specific example of a planar model. A specific example of a planar model. An example (a) of setting alpha values for the vertices of polygons constituting a planar model, and an example (b) of setting alpha values for each pixel of the planar polygon. The figure explaining the process which renders a 1st rendering image to a planar model. The figure which has arrange | positioned or displayed the planar model with which the 1st rendering image was affixed. The flowchart of an image process. The figure explaining the process by which a several 1st rendering image is arrange | positioned continuously, without connecting. The figure explaining the process by which the end of a some 1st rendering image is connected, and is arrange | positioned continuously.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Control part 11 CPU
12 signal processor 13 image processor 20 storage unit 21 RAM
22 interface unit 23 storage medium 30 image display unit 31 monitor 32 interface unit 40 audio output unit 41 interface unit 42 digital / analog conversion unit 43 amplifier 44 speaker 50 operation input unit 51 interface unit 52 operation unit 60 communication unit 70 bus 80, 90 , 110, 120 planar model 81-84, 91-92, 111, 121-124 polygon 801-803 planar model 820 reference object 1001, 1002 three-dimensional model

Claims (9)

First rendering means for determining a position and an angle of a three-dimensional model and rendering at a first camera position to generate a plurality of first rendered images;
A model reading unit that is a planar model composed of one or a plurality of polygons and that reads a plurality of planar models corresponding to each first rendering image from the storage unit;
With paste on each planar model corresponding to each of the first rendering image, and set the transparency of at least a portion of the vertexes of the polygons forming the planar model corresponding to each of the first rendering image, the apex Second rendering means for setting the transparency of each pixel on the planar model based on the transparency, assigning the transparency set for each pixel to each pixel of the first rendering image, and rendering;
Each planar model disposed in accordance with the position and angle of the three-dimensional model corresponding to each of the first rendering image, rotate the respective planar models to be line of sight perpendicular of the first camera position And third rendering means for rendering the first rendering image attached to each planar model ,
The second rendering means has a first transparency displayed at a relatively late time, with the transparency set in the first rendered image displayed at a relatively early time among the plurality of first rendered images. An image processing program for setting the transparency so that it is larger than the transparency set for the rendered image .   The image processing program according to claim 1, wherein the second rendering unit sets the transparency of each pixel on the planar model based on the transparency set at the vertex and the distance from the vertex to each pixel. . The image processing program according to claim 2 , wherein the second rendering unit changes the shape of the planar model by changing the positions of at least some of the vertices of the polygon. The image processing program according to claim 2 , wherein the model reading unit sets the plurality of planar models by selecting a plurality of planar models from a plurality of planar models stored in advance. The second rendering means positions said plurality of planar model so connected that one end of the plurality of first rendering image, claims 1 to 3 or according image processing program. The image processing program according to any one of claims 1 to 5 , wherein the second rendering means pastes a first rendered image rendered in any frame before the current frame to a plurality of planar models of the current frame. . The image processing program according to any one of claims 1 to 6 , wherein the second rendering unit changes the transparency set to the vertex of the polygon of the planar model for each frame. A first rendering means for determining a position and an angle of the three-dimensional model and rendering at the first camera position to generate a plurality of first rendered images;
A model reading unit that reads a plurality of planar models each composed of one or a plurality of polygons corresponding to each first rendering image ;
Second rendering means, with paste on each planar model corresponding to each of the first rendering image, setting the transparency in at least part of the vertexes of the polygons forming the planar model corresponding to each of the first rendering image and a step of rendering assigned based on the transparency of the vertices to set the transparency of each pixel on the planar model, the set transparency to each pixel to each pixel of said first rendering image,
Third rendering unit, according to the position and angle of the three-dimensional model corresponding to each of the first rendering image place each planar model, the first camera position of the line of sight and the plane so as to be perpendicular Rotating the shape model to render a first rendered image attached to each planar model ,
The second rendering means is a first rendering in which the transparency set in the first rendering image displayed at a relatively early time among the plurality of first rendering images is displayed at a relatively later time. An image processing method for setting the transparency so as to be larger than the transparency set for the image. First rendering means for determining a position and an angle of a three-dimensional model and rendering at a first camera position to generate a plurality of first rendered images;
Model reading means for reading out a plurality of planar models corresponding to each first rendering image from one or more planar models from the storage means;
With paste on each planar model corresponding to each of the first rendering image, and set the transparency of at least a portion of the vertexes of the polygons forming the planar model corresponding to each of the first rendering image, of the vertices Second rendering means for setting the transparency of each pixel on the planar model based on the transparency, assigning the transparency set for each pixel to each pixel of the first rendering image, and rendering;
The planar model disposed in accordance with the position and angle of the three-dimensional model corresponding to each of the first rendering image, rotate the respective planar models to be line of sight perpendicular of the first camera position And third rendering means for rendering the first rendering image attached to each planar model ,
The second rendering means has a first transparency displayed at a relatively late time, with the transparency set in the first rendered image displayed at a relatively early time among the plurality of first rendered images. An image processing apparatus for setting transparency so as to be larger than transparency set for a rendering image .

Images