JP5085722B2 - Image generation system - Google Patents

Description

  The present invention relates to a method for producing a stereoscopic print and a stereoscopic print.

Conventionally, a left-eye image taken with a camera corresponding to the left eye and a right-eye image taken with a camera equivalent to the right eye are prepared, and these images are synthesized by anaglyph processing or the like for stereoscopic viewing. A technique for obtaining an image for use (stereoscopic print) is known.

JP 2000-56411 A

  Now, humans feel the three-dimensional effect of an object. (1) Binocular parallax (gaze angle shift) in which the retina image is shifted due to the spatial separation of the left and right eyes, (2) This is caused by three physiological functions: convergence, which is a function in which the left and right eyes face inward, and (3) focus adjustment (focal length) in which the lens thickness responds to the distance to the object. Humans feel a three-dimensional feeling by processing these three physiological functions, binocular parallax, convergence, and focus adjustment, in the brain.

  The relationship between these three physiological functions is usually related in the brain. Therefore, if an error or contradiction arises in this relationship, the brain may forcibly associate with the solid, causing a situation where it feels unnatural or cannot be recognized as a solid.

  However, in conventional stereoscopic vision, stereoscopic vision is expressed using only binocular parallax and convergence. For this reason, the focus (focal length) is almost constant in the plane of the stereoscopic image (stereoscopic printed matter), whereas binocular parallax and convergence shift occur in almost all places of the stereoscopic image. , I could not realize a reasonable stereoscopic view in the human brain.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a method for manufacturing a stereoscopic print and a stereoscopic print that can realize more natural stereoscopic vision. is there.

  The present invention projects the left eye image by projecting each point of the object onto the reference plane that is non-orthogonal to the viewing direction in the projection direction connecting the left eye viewpoint position and each point of the object. Create a right-eye image by projecting and rendering each point of the object on the reference plane that is non-orthogonal to the line-of-sight direction in the projection direction connecting the right-eye viewpoint position and each point of the object. In addition, the present invention relates to a method for manufacturing a stereoscopic printed material that creates a stereoscopic printed material based on a left-eye image and a right-eye image.

  According to the present invention, the image for the left eye is created by projecting and rendering each point of the object on the reference plane in the projection direction connecting the viewpoint position for the left eye and each point of the object. Further, the right eye image is created by projecting and rendering each point of the object on the reference plane in the projection direction connecting the right eye viewpoint position and each point of the object. Based on these left-eye image and right-eye image, a stereoscopic print is created. Thereby, there is little contradiction of focus adjustment and a feeling of depth, and the printed matter for stereoscopic vision that can realize more natural stereoscopic vision can be provided. The reference plane is, for example, a plane that is not orthogonal to the line-of-sight direction (the direction connecting the midpoint of the left-eye viewpoint position and the right-eye viewpoint position and the gazing point of the virtual camera). In other words, it is a surface different from the perspective conversion screen orthogonal to the line-of-sight direction.

  In the present invention, the object projected onto the reference plane may be an object arranged on the reference plane. In the present invention, the object projected onto the reference plane may be an object in which part or all of the object is disposed on the back side of the reference plane when viewed from the left eye viewpoint position and the right eye viewpoint position. .

  In the present invention, the reference plane includes a first reference plane and a second reference plane that forms a predetermined angle with respect to the first reference plane, and a projection direction that connects the left eye viewpoint position and each point of the object. Then, each point of the object is projected onto the first or second reference plane and rendered on the first or second reference plane to create a left-eye image, and the right-eye viewpoint position and each of the object The right eye image may be created by projecting each point of the object onto the first or second reference plane and rendering the first or second reference plane in the projection direction connecting the points. Good.

  In this way, it is possible to solve this problem even when unnaturalness occurs in the sense of depth, etc., by setting only one reference plane. Three or more reference planes may be set. A plurality of reference planes (first and second reference planes) can be connected.

  In the present invention, when the distance between the object and the viewpoint position is increased, the distance between the left-eye viewpoint position and the right-eye viewpoint position may be increased according to the change in the length. .

  The viewpoint position is, for example, the midpoint between the left-eye viewpoint position and the right-eye viewpoint position.

  In the present invention, when the distance between the object and the viewpoint position is changed, the viewpoint position may be moved along a straight line that forms a predetermined angle with respect to the reference plane.

  In the present invention, the surface on which the stereoscopic printed material is placed during stereoscopic viewing may be set as the reference surface.

  In this way, when the stereoscopic printed material is placed, for example, parallel to the placement surface, it is possible to realize optimal and realistic stereoscopic viewing.

  In the present invention, the stereoscopic print may be created by combining the left-eye image and the right-eye image by anaglyph processing.

  In the present invention, the stereoscopic print may be created by a method other than the anaglyph process.

  Moreover, this invention relates to the printed matter for stereoscopic vision created by any one of the manufacturing methods described above. The present invention also relates to a stereoscopic printed matter created by duplicating a stereoscopic printed matter created by any one of the above-described manufacturing methods.

It is a flowchart of the 1st stereoscopic vision system of this embodiment. It is explanatory drawing of the 1st stereoscopic vision system of this embodiment. It is an example of image IL1 for left eyes. It is an example of the image for right eye IR1. It is an example of image IL2 for left eyes. It is an example of the image for right eye IR2. It is an example of the image for stereoscopic vision (anaglyph image). 8A, 8B, and 8C are explanatory diagrams of correction processing that eliminates the perspective. It is explanatory drawing of the characteristic of the image for stereoscopic vision obtained by this embodiment. It is explanatory drawing of the method of providing a some reference surface. It is a flowchart of the method of providing a some reference plane. It is explanatory drawing of the method of providing a some reference surface. It is explanatory drawing of the 2nd stereoscopic vision system of this embodiment. 14A, 14B, and 14C are explanatory diagrams of the second stereoscopic viewing method. 15A and 15B are explanatory diagrams of a conventional method. It is explanatory drawing of the setting method of a viewpoint position. It is explanatory drawing of the production method of the printed matter for stereoscopic vision using a real image. It is explanatory drawing of the production method of the printed matter for stereoscopic vision using a real image. 19A and 19B are explanatory diagrams of a method for creating a stereoscopic printed matter using a CG image. It is explanatory drawing of the correction | amendment process using texture mapping. 1 is a configuration example of an image generation system. It is explanatory drawing about a coordinate system. It is explanatory drawing about the conversion from G1-G4 to G1'-G4 ', and the conversion from G1'-G4' to F1-F4. It is a figure for deriving the conversion formula of the 2nd stereoscopic vision system. FIGS. 25A and 25B are explanatory diagrams of conversion constituting the second stereoscopic viewing system. It is explanatory drawing of the conversion which comprises a 2nd stereoscopic vision system. It is explanatory drawing about the deviation | shift amount in conversion of a 1st stereoscopic vision system.

  Hereinafter, the present embodiment will be described with reference to the drawings.

  In addition, this embodiment demonstrated below does not unduly limit the content of this invention described in the claim. In addition, all the configurations described in the present embodiment are not necessarily essential configuration requirements of the present invention.

  In the present embodiment, stereoscopic vision is realized by the two methods described below.

1. First Stereoscopic Method FIG. 1 shows a flowchart of a first stereoscopic method according to this embodiment.

  First, a first left-eye image IL1 and a first right-eye image IR1 for stereoscopic viewing are created (generated) (steps S1 and S2). Specifically, a left eye image IL1 seen from the left eye viewpoint position VPL and a right eye image IR1 seen from the right eye viewpoint position VPR are created (generated).

  Here, the left-eye and right-eye viewpoint positions VPL and VPR are positions assumed as positions of the left eye and right eye of the viewer, as shown in FIG. For example, when the left-eye and right-eye images IL1 and IR1 are created by actual shooting with a camera (digital camera), the left-eye and right-eye images IL1 and IR1 are obtained by arranging the cameras at these VPL and VPR positions. Take a picture. In this case, two cameras may be arranged in the VPL and VPR and photographed at the same time, or the position of one camera may be changed and photographed.

  On the other hand, when the left-eye and right-eye images IL1 and IR1 are generated by a system that generates CG (computer graphics) images and game images (real-time moving images), virtual cameras are arranged at these VPL and VPR positions. Thus, the left-eye and right-eye images IL1 and IR1 are generated. That is, an image that can be seen from the VPL and VPR in the object space is generated.

  3 and 4 show examples of the left-eye image IL1 and the right-eye image IR1. These are examples in the case where IL1 and IR1 are created by actual shooting with a camera (digital camera). Various objects such as mandarin oranges, boxes, ballpoint pens, and staplers (subjects or objects in a narrow sense; the same applies to the following description) are arranged on a reference surface (a placement surface on which an object is placed). The left-eye image IL1 is obtained by placing the camera at the left-eye viewpoint position VPL and taking the camera's line of sight (direction) toward the object (gaze point, representative point of the object). The right-eye image IR1 is obtained by placing a camera at the right-eye viewpoint position VPR and shooting with the camera's line of sight directed toward the object. As shown in FIGS. 3 and 4, the left-eye and right-eye images IL1 and IR1 have different viewing angles (appearance), and stereoscopic viewing is possible using binocular parallax due to the viewing angle deviation. Realized.

  In the present embodiment, the surface (desk or table surface; horizontal surface) on which the stereoscopic printed material is placed can be set as the reference surface during stereoscopic viewing.

  In the case of a CG or a game, an object (an object that models a mandarin orange, a box, a ballpoint pen, a stapler, etc.) is placed on a reference plane set in the object space, and a virtual camera is placed in the VPL or VPR. Deploy. Then, by generating an image that can be seen from the virtual camera with the line of sight (direction) of the virtual camera directed toward the object (gaze point, representative point of the object), it is possible to generate an image similar to that in FIGS.

  Next, as shown in step S3 of FIG. 1, the first left-eye image IL1 obtained in step S1 is subjected to correction processing for eliminating the perspective of the image on the reference plane BS, and the second The left eye image IL2 is created (generated). Further, as shown in step S4, the first right-eye image IR1 obtained in step S2 is subjected to correction processing for eliminating the perspective of the image on the reference plane BS, and the second right-eye image is obtained. IR2 is created (generated).

  5 and 6 show examples of the left-eye image IL2 and the right-eye image IR2 obtained by the correction process. For example, in FIGS. 3 and 4, a perspective is attached to a rectangle RTG drawn in the reference plane BS (a rectangle in a broad sense including a square, which is the same in the following description). On the other hand, in FIGS. 5 and 6, the perspective of the rectangular RTG is lost.

  Here, the correction process for eliminating the perspective in the present embodiment is, as shown in FIG. 8A, an image of the reference plane BS itself, an image IM1 drawn on the reference plane, and an object OB (object). This is a process for eliminating the perspective (depth feeling) of the image in the part of the image that is in contact with the reference plane BS. That is, in B1 of FIG. 8A, the distance between the vertices decreases as the distance from the viewpoint increases, but in B2 of FIG. 8A, the distance between the vertices changes even when the distance from the viewpoint increases. Absent. By performing such correction processing, an image as if viewed from directly above is created (generated) for the image of the reference surface BS. Note that the perspective does not need to be completely eliminated by this correction process, and it is sufficient that the perspective disappears to such an extent that a sense of incongruity does not occur.

  Next, as shown in step S5 of FIG. 1, a stereoscopic image (image data) is created (generated) based on the second left-eye image IL2 and the second right-eye image IR2. More specifically, a stereoscopic image is created (generated) by performing anaglyph processing or the like based on IL2 and IR2.

  Then, this stereoscopic image (actual image or CG image) is printed on a print medium by using a color printer (printer in a broad sense) such as an ink jet method or a laser printer method, so that a printed matter for stereoscopic vision is obtained. Can be manufactured. In addition, you may manufacture the printed material for stereoscopic vision by duplicating the printed material for stereoscopic vision used as the original disc printed by the color printer (printing machine). In this way, there is an advantage that a large amount of stereoscopic printed material can be manufactured in a short period of time.

  If the stereoscopic image is displayed on the display unit of the image generation system, a game image (moving image) can be generated in real time. In this case, a stereoscopic image obtained by anaglyph processing or the like is directly displayed on the display unit, and this is viewed using glasses (equipment in a broad sense) provided with color filters (red and blue). You may do it. Alternatively, the left-eye and right-eye images IL2 and IR2 may be alternately displayed on the display unit in different frames, for example, and viewed using glasses provided with a liquid crystal shutter or the like.

  FIG. 7 shows an example of a stereoscopic image obtained by performing anaglyph processing based on the left-eye and right-eye images IL2 and IR2 in FIGS.

  In the stereoscopic image of FIG. 7, the left-eye image IL2 (IL) and the right-eye image IR2 (IR) are synthesized. The left-eye image IL2 and the right-eye image IR2 each include an image of the object OB arranged on the reference plane BS. It also includes an image of the reference plane BS.

  As shown in A1 of FIG. 9, the object image of the left-eye image IL2 and the object image of the right-eye image IR2 match at the position of the reference plane BS (however, it is not always necessary to match completely). . That is, the printing position (display position) of the object image of the left-eye image IL2 and the printing position (display position) of the object image IR2 of the right-eye image are the same on the reference plane BS.

  On the other hand, as shown by A2 in FIG. 9, the distance between the object image of the left-eye image IL2 and the object image of the right-eye image IR2 increases as the distance from the reference plane BS increases. More specifically, the image of the portion of the object OB that is located above the reference plane BS is printed at the left eye image IL2 (display position) and at the right eye image IR2 (display position). ) Is off.

  A stereoscopic printed matter can be manufactured by printing the stereoscopic image shown in FIGS. 7 and 9 on a print medium. Then, stereoscopic viewing can be realized by viewing the printed material for stereoscopic viewing with glasses having a red filter on the left eye and a blue filter on the right eye, for example. Further, by displaying the stereoscopic image of FIGS. 7 and 9 on the display unit, it is possible to generate a stereoscopic game image.

  In the conventional stereoscopic vision, as shown in FIG. 8B, the plane of the stereoscopic printed matter PM (or the display screen of the display unit; the same applies to the following description) is parallel to the vertical plane. It was assumed that the viewer would see the stereoscopic printed matter PM facing each other. Therefore, for example, the left-eye and right-eye images IL1 and IR1 as shown in FIGS. 3 and 4 are subjected to anaglyph processing as they are to create a stereoscopic print PM. Since the perspectives remain in the images of FIGS. 3 and 4, when the stereoscopic printed material PM is viewed face to face as shown in FIG. 8B, the images are correct as far as perspective is concerned.

  However, when the viewer looks at the stereoscopic printed matter PM as shown in FIG. 8B, the focus (focal length) is the same on the entire surface of the PM. Accordingly, in the human brain, contradiction and error occur in the relationship between focus adjustment, binocular parallax, and convergence. Therefore, when the brain tries to associate with a solid forcibly, it feels unnatural or cannot be recognized as a solid. Further, if the stereoscopic printed matter PM created by the conventional method is arranged on a desk so as to be parallel to a horizontal plane, the depth feeling becomes inconsistent, resulting in an unnatural stereoscopic view. That is, the rectangular RTG in FIGS. 3 and 4 is a plane having a height of zero, and this rectangular RTG should not appear as a solid.

  Therefore, in this embodiment, as shown in FIG. 8C, it is assumed that the viewer sees the stereoscopic printed matter PM (display screen) placed on a desk (reference plane BS parallel to the horizontal plane). Like to do. That is, such an arrangement is the default arrangement of this method. When the stereoscopic printed matter PM is arranged in parallel with the horizontal plane as described above, if the images of FIGS. 3 and 4 are anaglyph processed as they are to create the stereoscopic printed matter PM, a contradiction occurs in perspective.

  Therefore, in this embodiment, as described with reference to FIGS. 5, 6, and 8A, correction processing is performed to eliminate the perspective of the reference plane image. Then, anaglyph processing is performed based on the corrected images of FIGS. 5 and 6 after eliminating the perspective on the reference plane, and a stereoscopic printed matter PM is created. The created stereoscopic printed matter PM is shown in FIG. If the reference plane image (rectangular RTG) is arranged parallel to the horizontal plane as in (), an appropriate perspective can be obtained. Moreover, if it arrange | positions like FIG.8 (C), the focal distance of each point on the surface of the printed matter PM for stereoscopic vision will become different instead of being the same. For this reason, the focus adjustment is similar to that in the real world. Therefore, the shift in the relationship between the focus adjustment and the binocular parallax and the convergence is reduced, and a more natural and realistic stereoscopic vision can be realized.

  Note that, in the stereoscopic viewing method of the present embodiment, when the height of the object is high, there may be a shift in the depth feeling or the like. In such a case, for example, as shown in FIG. 10, two reference surfaces BS1 and BS2 (a plurality of reference surfaces in a broad sense) may be provided.

  Here, the reference plane BS1 is, for example, a plane parallel to the horizontal plane. On the other hand, the reference surface BS2 is a surface that forms a predetermined angle (for example, a right angle) with the reference surface BS1. The reference surfaces BS1 and BS2 are connected at the boundary BD.

  The object OB (object) is arranged above the reference surface BS1 and on the front side (VPL, VPR side) of the reference surface BS2. Then, the processing shown in FIG. 11 is performed instead of FIG.

Steps S11 and S12 in FIG. 11 are the same as steps S1 and S2 in FIG. In step S13, correction processing for eliminating the perspective on the reference plane BS1 is performed on the area corresponding to the reference plane BS1 of the left-eye image IL1 (the first area on the BS1 side of IL1 with reference to the boundary BD). Apply to. Further, correction processing for eliminating the perspective on the reference surface BS2 is performed on the region corresponding to the reference surface BS2 of IL1 (the second region on the BS2 side of IL1 with reference to the boundary BD). Then, a left-eye image Il2 that is an image obtained by connecting the images generated by these correction processes is created (generated).

  In step S14, correction processing for eliminating the perspective on the reference surface BS1 is performed on the region corresponding to the reference surface BS1 of the right-eye image IR1 (the first region on the BS1 side of the IR1 with reference to the boundary BD). Apply to. Further, correction processing for eliminating the perspective on the reference surface BS2 is performed on the region corresponding to the reference surface BS2 of IR1 (the second region on the BS2 side of IR1 with reference to the boundary BD). Then, a right-eye image IR2 that is an image obtained by connecting the images generated by these correction processes is created (generated).

  Finally, as in step S15, based on IL2 and IR2, for example, an anaglyph process is performed to create (generate) a stereoscopic image. Then, the obtained stereoscopic image is printed on a printing medium to produce a stereoscopic printed material, or the stereoscopic image is displayed on the display unit to generate a game image that is a real-time moving image.

  By doing so, as shown in FIG. 12, even when the OB is an object having a high height from the reference plane BS1, a more natural and realistic stereoscopic vision can be realized. That is, in the region near the foot of the object OB (the first region below the boundary BS), stereoscopic processing using the reference plane BS1 can realize stereoscopic viewing that does not make sense of depth and focus adjustment. On the other hand, in other regions (second region on the upper side of the boundary BS), stereoscopic viewing without a sense of depth can be realized by the stereoscopic processing using the reference plane BS2.

  The reference plane is not limited to two, and three or more reference planes (a plurality of linked reference planes) may be used.

2. Second Stereoscopic Viewing Method FIG. 13 shows a flowchart of the second stereoscopic viewing method of the present embodiment. The above-described method of FIG. 1 is an optimal method for creating a stereoscopic print using an image captured by a camera, whereas the method of FIG. 13 uses a CG image to provide a stereoscopic print. Is the best way to create

  First, in the projection direction connecting the left eye viewpoint position VPL and each point of the object OB, each point of OB is projected on the reference surface BS (BS1 or BS2 in the case of FIG. 10) and rendered on the reference surface BS, and the left eye An image IL is created (generated) (step S21).

  Next, in the projection direction connecting the right eye viewpoint position VPR and each point of the object OB, each point of OB is projected onto the reference plane BS (BS1 or BS2 in the case of FIG. 10) and rendered on the reference plane BS, and the right eye A production image IR is created (generated) (step S22). Note that the reference surface BS is a surface that is not orthogonal to, for example, the line-of-sight direction (the direction connecting the viewpoint position and the gazing point). That is, the reference plane BS is a plane different from the perspective projection screen that is always orthogonal to the line-of-sight direction.

  In the processing of steps S21 and S22, virtual light is projected from the VPL (or VPR) toward the object OB, and the image of the OB is a virtual that is the reference plane BS (BS1 or BS2) using the light. Render to virtual paper as if it were burned on paper. Thereby, as shown in FIG. 14A, the images (properties such as colors) of the points P1, P2, P3, and P4 of the object OB are projected points P1 ′, P2 ′, P3 ′, Rendered to P4 ′. Note that the images of the points P5 and P6 on the reference plane BS are rendered as they are at the positions of the points P5 and P6. Then, for example, as shown in FIG. 14B, rendering is performed so that the entire surface of the reference surface BS (virtual paper) is raster-scanned, whereby the left-eye image IL, the right-eye image similar to IL2 and IR2 in FIGS. An image IR can be created. That is, it is possible to create the left-eye and right-eye images IL and IR without the perspective of the reference plane image.

  Then, based on these left-eye and right-eye images IL and IR, for example, an anaglyph process is performed to create (generate) a stereoscopic image (step S23). Thereby, a stereoscopic image as shown in FIG. 7 can be obtained. Then, the obtained stereoscopic image is printed on a print medium to produce a stereoscopic print, or the game image can be generated by displaying the stereoscopic image on the display unit.

  For example, as shown in FIG. 14C, the stereoscopic printed matter PM (or display screen) is arranged so as to be parallel to the horizontal plane (reference plane), so that a more natural and realistic stereoscopic view can be obtained. realizable.

  For example, in FIG. 15A, an object OB is perspectively projected on a perspective projection screen SCR (a plane orthogonal to the line-of-sight direction) to create a left-eye image and a right-eye image. Then, the obtained left-eye image and right-eye image are synthesized to create a stereoscopic printed matter PM. Then, as shown in FIG. 15B, the viewer sees the PM facing the stereoscopic printed material PM.

  In the method of FIG. 15A, the points P2 and P3 of the object OB are projected onto the points P2 ″ and P3 ″ on the projection projection screen SCR. Then, since the stereoscopic printed matter PM is viewed directly facing as shown in FIG. 15B, the focal length difference L2 between P2 ″ and P3 ″ becomes zero. That is, since the focal length difference L1 between the actual points P2 and P3 is not 0 but L2 is 0, the focus adjustment is different from the actual focus adjustment. Accordingly, a contradiction arises in the relationship between the focus adjustment and the binocular parallax, and the human brain is confused, resulting in an uncomfortable stereoscopic view.

  On the other hand, in the present embodiment, the stereoscopic printed matter PM (display screen) is placed on a desk as shown in FIG. 14C, and therefore, as shown in FIG. The focal length difference L2 of ', P3' is not 0, like the focal length difference L1 of the actual points P1, P2. Therefore, the front part (point P2) can be seen in the front, and the back part (P3) can be seen in the back, so there is no contradiction between the focus adjustment and the binocular parallax, and the human brain is confused. Therefore, more natural stereoscopic vision can be realized.

  That is, in this embodiment, since the stereoscopic printed matter PM is placed on a desk and viewed from an oblique direction, the desk surface and the reference plane BS (zero plane) on which the object OB to be stereoscopically viewed is placed are described. It becomes the same surface, is realistic, and does not cause unreasonable stereoscopic vision. The object OB only needs to be able to express the appearance that the object OB is lifted by a few centimeters with respect to the reference plane BS (zero plane), so that there is almost no contradiction in the depth direction. In addition, since the reference surface BS is a desk surface, it looks as if a three-dimensional object is actually placed on the desk, and the realism of the object is improved. That is, in the conventional method shown in FIGS. 15A and 15B, the reference surface is dim, so that there is a certain three-dimensional effect, but the real sense of the object has only become a phantom.

  In the method of FIG. 13, as described with reference to FIG. 10, a plurality of reference planes may be set to create (generate) a stereoscopic image. In this case, in steps S21 and S22 of FIG. 13, the points projected on the reference plane BS1 may be rendered on the reference plane BS1, and the points projected on the reference plane BS2 may be rendered on the reference plane BS2.

3. Anaglyph Processing Next, the anaglyph processing performed in step S5 in FIG. 1, step S15 in FIG. 11, and step S23 in FIG. 13 will be briefly described.

In the anaglyph process, the left-eye image and the right-eye image are printed on a single print medium with different colors to create a stereoscopic print. The stereoscopic print is viewed through different color filters (for example, the left eye is red and the right eye is blue) between the left and right eyes. At this time, only the image for the left eye can be seen with the left eye, and only the image for the right eye can be seen with the right eye, thereby realizing stereoscopic viewing.

  For example, in monochrome anaglyph processing, the image for the left eye (IL2, IL) is converted to gray scale. Then, the converted image data is copied to the R channel of the anaglyph image (RGB). Next, the image for the right eye (IR2, IR) is converted to gray scale. Then, the converted image data is copied to the G channel and B channel of the anaglyph image (RGB). Thereby, a monochrome anaglyph image is created. Note that the right-eye image may be copied only to the B channel.

  In color anaglyph processing, the R channel of the left-eye image (IL2, IL) is copied to the R channel of the anaglyph image (RGB). The G channel of the right eye image (IR2, IR) is copied to the G channel of the anaglyph image (RGB). Also, the B channel of the right eye image is copied to the B channel of the anaglyph image (RGB). Thereby, a color (pseudo-color) anaglyph image can be created.

  Note that the method for realizing stereoscopic vision (step S5 in FIG. 1, step S15 in FIG. 11, and step S23 in FIG. 13) uses at least a left-eye image (IL2, IL) and a right-eye image (IR2, IR). What is realized is not limited to anaglyph processing.

  For example, using a special lens called a lenticular lens, stereoscopic vision may be realized such that only the image for the left eye enters the left eye and only the image for the right eye enters the right eye.

  In addition, a polarizing plate is disposed in front of the left-eye image and the right-eye image, and the polarizing direction is different between the polarizing plate placed in front of the left-eye image and the polarizing plate placed in front of the right-eye image. . Then, stereoscopic viewing may be realized by the viewer wearing spectacles in which a polarizing plate having a deflection direction corresponding thereto is attached to the lens portion.

  Also, the left-eye image and the right-eye image are displayed alternately, for example, with different frames. Then, the viewer wears glasses equipped with a shutter for the left eye (for example, a liquid crystal shutter) that opens in synchronization with the display of the image for the left eye and a shutter for the right eye that opens in synchronization with the display of the image for the right eye. Visualization may be realized.

4). Setting the viewpoint position Next, a method for setting the viewpoint position will be described.

  The left-eye and right-eye viewpoint positions VPL and VPR in FIGS. 2 and 10 are arranged based on the assumed positions of the left and right eyes of the viewer when the viewer actually views the stereoscopic print or the stereoscopic display screen. It is desirable to do. For example, in FIGS. 2 and 10, the distance DVB (for example, 40 cm) between the object OB (object, subject) and the viewer's eyes, the viewing angle θ (the viewing direction SL), and the distance DLR (for example, 7 cm) between the eyes. Based on this, the left-eye and right-eye viewpoint positions VPL and VPR are set.

  However, when performing reduction display or enlargement display, the positions of VPL and VPR are moved according to the reduction ratio and the enlargement ratio. In this case, it is desirable to move the viewpoint position by a method as shown in FIG.

For example, when the distance DVB between the object OB (subject, object) and the viewpoint position (middle point CP of VPL and VPR) is increased, the left-eye viewpoint position V according to the change (ratio) of the length.
The distance DLR between the PL and the right eye viewpoint position VPR is increased. That is, for example, the DLR is lengthened in proportion to the change in the length of DVB.

  Further, when changing the distance DVB between the object OB (subject, object) and the viewpoint position (middle point CP of VPL and VPR), a straight line LN (line-of-sight direction) forming a predetermined angle θ with respect to the reference plane BS. The viewpoint position (midpoint CP, VPL, VPR) is moved so as to move along.

  In this way, even when the VPL and VPR are moved, the distance DVB and the distance DLR change at the same magnification ratio, so that it is possible to prevent a situation in which the stereoscopic effect is broken. Thereby, reduction display and enlargement display can be realized while maintaining an appropriate stereoscopic effect.

5. Next, details of a method of creating (manufacturing) a stereoscopic print using a real image will be described. In this case, the first stereoscopic method described with reference to FIG. 1 is suitable.

  When using a photographed image, it is necessary to reproduce the environment at the time of photographing as it is. Accordingly, a camera for actual shooting (such as a digital camera) is arranged in a layout close to the positional relationship when the viewer looks. For example, a stereoscopic print is placed on a standard desk, and a live-action camera is arranged assuming that the viewer is sitting on a chair and watching.

5.1 When there is a single reference plane When there is a single reference plane as shown in FIG. 2, the distance DLR (approximately 7 cm) between both eyes, the distance DVB between the viewpoint and the subject OB, the viewing angle θ, the printing The vertical size D1 and horizontal size D2 (printing range) of the paper are set.

  Next, a camera is arranged at a position assumed to be the position of the left eye and right eye of the viewer. Then, paper on which marks MK1 to MK4 (first to fourth marks) serving as a guide for the printing range (D1, D2) are written is placed. The marks MK1 to MK4 constitute the vertices of a rectangle (a broad rectangle including a square) on the reference plane BS.

  Next, the subject OB is placed on paper. At this time, the OB is placed so that the subject OB does not protrude outside the rectangle (printing range) constituted by the marks MK1 to MK4 when viewed from the camera position. Then, using a camera set at a position assumed to be the position of the left eye and the right eye, the subject OB and the marks MK1 to MK4 are photographed so that the left eye and the right eye as shown in FIGS. Images IL1 and IR1 are created.

  Next, the captured left-eye and right-eye images IL1 and IR1 are loaded into an image generation system (personal computer, information processing apparatus) and displayed on the screen. Then, paper marks MK1 to MK4 are found from the displayed image.

  Next, as shown in FIG. 17, correction is performed by moving the marks MK1 to MK4 to the positions of the vertices VX1 to VX4 of the rectangle having the aspect ratio of D1 to D2 (a rectangle in a broad sense including a square) to distort the image. Process. By performing this correction process on each of the left-eye and right-eye images IL1 and IR1, left-eye and right-eye images IL2 and IR2 as shown in FIGS. 5 and 6 are created.

Next, an extra portion outside the printing range is trimmed. Then, using the anaglyph processing software, a stereoscopic image (anaglyph image) as shown in FIG. 7 is created from the left-eye and right-eye images IL2 and IR2. Then, the obtained stereoscopic image is printed on paper in a printing range having a size of D1 and D2, thereby completing a stereoscopic print.

5.2 When there are two reference planes When two reference planes are provided as shown in FIG. 10, the distance DLR (approximately 7 cm) between the eyes, the distance DVB between the viewpoint and the subject OB, the angle of view θ, the printing paper Vertical size D1, horizontal size D2, and height size D3 (printing range) are set.

  Next, cameras are arranged at positions assumed to be the positions of the left eye and right eye of the viewer. Then, the first sheet of paper (reference surface BS1) on which marks MK1 to MK4 (first to fourth marks) serving as guidelines for the printing range (D1, D2) are written is placed. The marks MK1 to MK4 constitute rectangular vertices on the reference plane BS1.

  Next, the second sheet of paper (reference surface BS2) on which marks MK5 to MK8 (fifth to eighth marks) serving as a guide for the printing range (D2, D3) are written is placed behind the first sheet. Stand vertically. The marks MK5 to MK8 constitute rectangular vertices on the reference plane BS2.

  Next, the subject OB is placed on the first sheet. At this time, the OB is placed so that the subject OB does not protrude outside the rectangle composed of the marks MK1 to MK4 and the rectangle composed of the marks MK5 to MK8 (printing range) when viewed from the camera position. Then, using a camera set at a position assumed to be the position of the left eye and the right eye, the subject OB and the marks MK1 to MK4 and MK5 to MK8 are photographed, and the left and right eye images IL1 and IR1 ( Photo).

  Next, the captured left-eye and right-eye images IL1 and IR1 are taken into an image generation system (personal computer) and displayed on the screen. Then, paper marks MK1 to MK4 and MK5 to MK8 are found from the displayed images. The marks MK3, MK4 and MK6, MK5 can be the same mark.

  Next, as shown in FIG. 18, correction processing is performed to distort the image by moving the marks MK1 to MK4 to the positions of the vertices VX1 to VX4 of the rectangle having the aspect ratio of D1 to D2. Further, correction processing is performed to distort the image by moving the marks MK5 to MK8 to the positions of the vertices VX5 to VX8 of the rectangle having the aspect ratio of D3 to D2. Then, the obtained two images are connected. By performing this correction process on each of the left-eye and right-eye images IL1 and IR1, left-eye and right-eye images IL2 and IR2 are created.

  Next, an extra portion outside the printing range is trimmed. Then, a stereoscopic image (anaglyph image) is created from the left-eye and right-eye images IL2 and IR2 using anaglyph processing software. Then, the obtained stereoscopic image is printed on paper in a printing range having a size of D1, D2, and D3 to complete a stereoscopic print.

6). Creation of Stereoscopic Print Using CG Image Next, a method for creating (manufacturing) a stereoscopic print using a CG (computer graphics) image will be described. In this case, the second stereoscopic method described with reference to FIG. 13 is suitable. However, it can also be realized by the first stereoscopic method of FIG.

  First, a virtual camera (viewpoint) is arranged in the object space with a layout close to the positional relationship when the viewer looks. For example, a virtual camera is arranged on the assumption that a stereoscopic print or the like is placed on a standard desk and a viewer sits on a chair.

Then, as shown in FIG. 2, the distance DLR (about 7 cm) between the eyes, the distance DVB between the viewpoint and the object OB, the line-of-sight angle θ, the vertical size D1 of the printing paper, and the horizontal size D2 (printing range) are set.

  Next, virtual cameras are arranged at positions assumed to be the positions of the left eye and right eye of the viewer. Then, the object is arranged on the virtual paper (virtual paper object).

  Next, virtual light is projected from the left-eye viewpoint position VPL toward the object OB, and rendering is performed using the light so that the image of the OB is printed on virtual paper. Thereby, the image for left eye IL is created. This is a process similar to the process of projecting an image viewed from the eye onto virtual paper on a desk by a projector.

  Next, virtual light is projected from the right eye viewpoint position VPR toward the object OB, and rendering is performed using the light so that the image of the OB is printed on virtual paper. As a result, a right-eye image IR is created.

  Next, a stereoscopic image (anaglyph image) is created from the left-eye and right-eye images IL and IR using anaglyph processing software. Then, the obtained stereoscopic image is printed on paper in a printing range having a size of D1 and D2, thereby completing a stereoscopic print.

  Note that a plurality of reference planes may be provided as shown in FIG. 10 to create a stereoscopic print using a CG image.

  Further, the object OB projected onto the reference plane BS may be an object arranged on the reference plane BS as shown in FIG. 2, or as shown in FIG. A part of the object may be an object disposed on the back side of the reference plane BS. Alternatively, as shown in FIG. 19B, all of the objects may be objects arranged on the back side of the reference plane BS.

  For example, in FIG. 19A, points P1, P2, and P3 on the back side of the reference plane BS are projected onto the points P1 ', P2', and P3 'in front. Thereby, a hole opened in the object OB can be expressed. In addition, at the position C1 in FIG. 19A, it is possible to express a state in which the object OB is embedded in the virtual paper.

  Also in FIG. 19B, the points P1, P2, and P3 on the back side of the reference plane BS are projected onto the points P1 ', P2', and P3 'in front. Thereby, an object such as a fish submerged under the water surface can be expressed. When expressing a semi-transparent object such as a water surface, a semi-transparent object is arranged at the position of the reference surface BS, and α synthesis of the semi-transparent object and the object OB (points P1 ′, P2 ′, P3 ′) is performed. It is desirable to do.

  As described above, according to the method of the present embodiment using a CG image, it is possible to create a stereoscopic printed matter that is optimal for attachment to a game manual.

  For example, in a conventional game manual in which only a planar map picture is attached, there is a problem that it is difficult for the player to grasp the shape of the map.

On the other hand, if the method of this embodiment is used, it becomes possible to attach the printed matter of the map which looks three-dimensionally to a game manual. For example, map shape data exists as game data, and therefore, using this game data, a map stereoscopic print can be easily created. In addition, according to the method of the present embodiment, it is possible to provide a stereoscopic printed material that gives the clearest stereoscopic effect when placed on a desk or the like. Therefore, it is possible to provide a stereoscopic printed matter that is easy to use and convenient for the player and is optimal for attachment to a game manual.

  Note that, for example, in a car, tank, airplane game, or the like, a stereoscopic printed material in which the appearing cars, tanks, and airplanes are three-dimensionally displayed may be attached to the game manual. Or if the method of this embodiment is applied to a monster card game, the card game in which the monster of a card | curd may jump out three-dimensionally is realizable. In particular, in a card game, a card is placed on a horizontal surface such as a desk or table to enjoy the game. Therefore, the method of the present embodiment that enables the most effective stereoscopic viewing when placed on a horizontal surface (reference surface) is optimal.

7). Generation of Game Image Next, a method for generating a game image that is a real-time moving image will be described. In this case, the first stereoscopic method described with reference to FIG. 1 is suitable. However, it can also be realized by the second stereoscopic method shown in FIG.

  First, a virtual camera (viewpoint) is arranged in the object space with a layout close to the positional relationship when the player views. For example, a virtual camera is placed on the assumption that a player is sitting on a chair and viewing a stereoscopic printed material on a standard desk.

  Then, as shown in FIG. 2, the distance DLR (about 7 cm) between the eyes, the distance DVB between the viewpoint and the object OB, the viewing angle θ, the vertical size D1 of the display screen, and the horizontal size D2 (display screen size) are set.

  Next, virtual cameras are arranged at the left-eye and right-eye viewpoint positions VPL and VPR that are assumed to be the positions of the left eye and right eye of the player. Also, an object OB that is a subject of the virtual camera is arranged in the object space. These virtual cameras are basically directed from the left-eye and right-eye viewpoint positions VPL and VPR toward the object (gazing object) in the object space.

  Next, left-eye and right-eye images IL1 and IR1 that are visible from the virtual cameras arranged at the left-eye and right-eye viewpoint positions VPL and VPR are generated. Then, the generated left-eye and right-eye images IL1 and IR1 are written in the texture area (texture space) of the VRAM, and these images are set to a texture image TEX as shown in FIG.

  Next, the perspective texture image TEX (see FIGS. 3 and 4) is mapped to a polygon PLG (primitive surface) of rectangles of D1 and D2 sizes (rectangles in a broad sense including squares). Specifically, the texture coordinates (TX1, TY1), (TX2, TY2), (TX3, TY3), (TX4, TY4) of the texture image TEX are coordinated with the vertices VX1, VX2, VX3, VX4 of the polygon PLG. Thus, the texture image TEX is mapped to the polygon PLG. As a result, as shown in FIGS. 6 and 7, it is possible to generate an image in which the perspective of the image of the reference plane is lost. Then, by performing such a texture mapping process for each of the left-eye and right-eye images IL1 and IR1, the left-eye and right-eye images IL2 and IR2 are generated.

  Next, the obtained left-eye and right-eye images IL2 and IR2 are combined into a single stereoscopic image using anaglyph processing. Then, the combined stereoscopic image is output to the display unit.

  In the case of realizing stereoscopic viewing using a liquid crystal shutter or the like, the generated left-eye and right-eye images IL2 and IR2 may be alternately output to the display unit in different frames.

8). Image Generation System FIG. 21 shows an example of a functional block diagram of the image generation system of this embodiment. Note that the image generation system of the present embodiment does not have to include all the components (each unit) in FIG. 21, and may have a configuration in which some of them are omitted.

  The image generation system of FIG. 21 can be used as a system for generating a game image (real-time moving image). Moreover, it can also be used as an image generation system (CG tool) for creating a stereoscopic image from a CG image (still image) and creating a stereoscopic print. Further, it can be used as an image generation system for taking a real image taken by a camera, creating a stereoscopic image from the real image, and creating a stereoscopic print.

  The operation unit 160 is for a player (operator) to input operation data, and functions thereof are a lever, a button, a steering, a shift lever, an accelerator pedal, a brake pedal, a microphone, a sensor, a touch panel, a casing, and the like. It can be realized by hardware.

  The storage unit 170 serves as a work area such as the processing unit 100 or the communication unit 196, and its function can be realized by hardware such as a RAM.

  The information storage medium 180 (computer-readable medium) stores programs, data, and the like, and functions as an optical disk (CD, DVD), magneto-optical disk (MO), magnetic disk, hard disk, and magnetic tape. Alternatively, it can be realized by hardware such as a memory (ROM). The processing unit 100 performs various processes of this embodiment based on a program (data) stored in the information storage medium 180. That is, the information storage medium 180 stores (records and stores) a program for causing the computer to function as each unit of the present embodiment (a program for causing the computer to execute processing of each unit).

  The display unit 190 outputs an image generated according to the present embodiment, and its function can be realized by hardware such as a CRT, LCD, touch panel, or HMD (head mounted display).

  The sound output unit 192 outputs the sound generated by the present embodiment, and its function can be realized by hardware such as a speaker or headphones.

  The portable information storage device 194 stores player personal data, game save data, and the like. Examples of the portable information storage device 194 include a memory card and a portable game device.

  The communication unit 196 performs various controls for communicating with the outside (for example, a host device or other image generation system), and functions thereof are hardware such as various processors or a communication ASIC, It can be realized by a program. Using this communication unit 196, it is possible to capture a real image captured by the camera into an image generation system and output the created stereoscopic image to a printer.

  Note that a program (data) for causing a computer to function as each unit of this embodiment is distributed from the information storage medium of the host device (server) to the information storage medium 180 (storage unit 170) via the network and the communication unit 196. You may do it. Use of such an information storage medium of the host device (server) is also included in the scope of the present invention.

The processing unit 100 (processor) performs various processes such as a game process, an image generation process, and a sound generation process based on operation data from the operation unit 160, a program, and the like. In this case, the processing unit 100 performs various processes using the main storage unit 172 in the storage unit 170 as a work area. The function of the processing unit 100 can be realized by hardware such as various processors (CPU, DSP, etc.) or ASIC (gate array, etc.) and a program (game program).

  The processing unit 100 includes a game processing unit 110, a first image generation unit 120, a second image generation unit 122, a stereoscopic image generation unit 126, and a sound generation unit 130.

  Here, the game processing unit 110 performs various game processes for generating a game image based on operation data from the operation unit 160 (game controller). This game process includes a process for starting a game based on game start conditions, a process for advancing the game, a process for placing an object (display object) appearing in the game, object movement information (position, speed, acceleration), There are a process for obtaining motion information (motion information), a process for displaying an object, a process for calculating a game result, a process for ending a game when a game end condition is satisfied, and the like.

  The first image generation unit 120 performs processing for generating a first left-eye image that is an image seen from the left-eye viewpoint position (left-eye virtual camera) in the object space. In addition, a process of generating a first right-eye image that is an image seen from the right-eye viewpoint position (right-eye virtual camera) in the object space is performed. In this case, the first left-eye image and the first right-eye image are images for stereoscopic viewing, for example, images with binocular parallax. Specifically, a virtual camera is arranged at the left-eye viewpoint position, and the first left-eye image is generated with the visual line direction of the virtual camera directed toward the object (gaze point). Also, a virtual camera is arranged at the right-eye viewpoint position, and a first right-eye image is generated with the visual line direction of the virtual camera directed toward the object (gaze point).

  Note that an image seen from a virtual camera can be generated as follows. That is, first, geometric processing such as coordinate transformation, clipping processing, perspective transformation or light source processing is performed, and based on the processing result, drawing data (primitive surface vertex position coordinates, texture coordinates, color data, normal vector or α value etc.). Then, based on this drawing data (primitive surface data), the object (one or a plurality of primitive surfaces) after perspective transformation (after geometry processing) is converted into image information in units of pixels such as a drawing buffer 174 (frame buffer, work buffer, etc.). Draw in a buffer that can be stored. As a result, an image that can be seen from the virtual camera in the object space is generated.

  The second image generation unit 122 performs a correction process for eliminating the perspective of the image on the reference plane with respect to the first left-eye image, and generates a second left-eye image. In addition, a correction process for eliminating the perspective of the image on the reference plane is performed on the first right-eye image to generate a second right-eye image (see FIGS. 1 and 8A).

  The correction process in this case is realized by the texture mapping unit 124 performing the texture mapping process described with reference to FIG. Specifically, the first left-eye image and the first right-eye image generated by the first image generation unit 120 are stored in the texture storage unit 176 as texture images. Then, the texture mapping unit 124 generates a second left-eye image by mapping the stored texture of the first left-eye image to a rectangular polygon. Further, the second right-eye image is generated by mapping the stored texture of the first right-eye image to a rectangular polygon.

The second image generation unit 122 may generate a left-eye image and a right-eye image based on the method described with reference to FIG. That is, the second image generation unit 122 generates the left-eye image by projecting and rendering each point of the object on the reference plane in the projection direction connecting the left-eye viewpoint position and each point of the object. Also, the right eye image is generated by projecting and rendering each point of the object on the reference plane in the projection direction connecting the right eye viewpoint position and each point of the object.

  The stereoscopic image generation unit 126 performs processing for generating a stereoscopic image based on the second left-eye image (left-eye image) and the second right-eye image (right-eye image). For example, the second left-eye image (left-eye image) and the second right-eye image (right-eye image) are synthesized by anaglyph processing to generate a stereoscopic image and output it to the display unit 190. In this case, for example, the player plays the game wearing glasses with red and blue color filters provided on the left and right eyes.

  Alternatively, the stereoscopic image generation unit 126 performs a process of outputting the second left-eye image (left-eye image) and the second right-eye image (right-eye image) to the display unit 190 in different frames, so that the stereoscopic image is displayed. May be realized. In this case, the player plays the game wearing glasses with shutters that open and close in synchronization with the frame.

  The sound generation unit 130 performs sound processing based on the results of various processes performed by the processing unit 100, generates game sounds such as BGM, sound effects, or sounds, and outputs the game sounds to the sound output unit 192.

  Note that the image generation system of the present embodiment may be a system dedicated to the single player mode in which only one player can play, or not only the single player mode but also a multiplayer mode in which a plurality of players can play. The system may also be provided.

  Further, when a plurality of players play, game images and game sounds to be provided to the plurality of players may be generated using one terminal, or connected via a network (transmission line, communication line) or the like. Alternatively, it may be generated using a plurality of terminals (game machine, mobile phone).

9. Analysis of First and Second Stereoscopic Viewing Methods Next, the first and second stereoscopic viewing methods of the present embodiment described with reference to FIGS. 1 and 13 are mathematically analyzed. The first stereoscopic viewing method is a photograph (FIGS. 3 and 4) in which an image of a real-world object that cannot be directly projected (C1) onto a reference plane (desk) screen is obtained by camera shooting (C2). This shows that the post-processing (C3. Processing for eliminating the perspective in FIG. 8A) can be reconfigured within a practical range. Therefore, mathematical analysis is performed on the difference between the first stereoscopic viewing method and the second stereoscopic viewing method in which a point of an object is projected directly onto a reference plane (desk) screen.

9.1 Mathematical Analysis of First Stereoscopic Viewing Method First, a viewpoint (v), a camera screen (s), an object, and a coordinate system for them are determined as shown in FIG. In FIG. 22, the point (x, y, z) of the object is projected onto the point (x ^* , y ^* ) on the screen (camera screen) by projection from the viewpoint.

  First, the camera photographing (C2) can be expressed by combining the matrix of the rotation Rx in the following equation (1) and the matrix of the projection Pz in the following equation (2).

Here, the matrix of the rotation Rx is a matrix for rotating the oblique line-of-sight direction so as to be parallel to the Z-axis direction. The matrix of the projection Pz is a matrix representing the projection from the viewpoint (Z = v) to the camera screen (Z = s). Α is an angle formed between the line-of-sight direction and the reference plane screen.

  Therefore, the camera photographing (C2) can be expressed as the following equation (3).

The above equation (3) can also be expressed by a conversion equation such as the following equation (4).

For example, as shown in FIG. 23, four grid points G1 = ^t (a, a, 0), G2 = ^t (−a, a, 0) forming a square on a reference plane (Z = 0) of a desk or the like. ), G3 = ^t (−a, −a, 0), G4 = ^t (a, −a, 0). “T” means transposition.

These lattice points G1 to G4 appear in lattice points G1 ′ to G4 ′ as shown in FIG. 23 by the conversion of the above equations (3) and (4). The coordinates of these lattice points G1 ′ to G4 ′ are expressed by the above equations (3) (4
) ^T (x, y, z), ^t (a, a, 0), (-a, a, 0), ^t (-a, -a, 0), ^t (a, -a, 0) By substituting, the following equations (5) to (8) are calculated.

In the post-processing of the first stereoscopic viewing method (C3. Processing for eliminating the perspective), these lattice points G1 ′ to G4 ′ are converted into lattice points F1 = ^t (b, b) constituting a two-dimensional square on the photograph. This is a projective transformation for F2 = ^t (−b, b), F3 = ^t (−b, −b), and F4 = ^t (b, −b). That is, it is projective transformation that maps the positions of the markers MK1 to MK4 (corresponding to G1 ′ to G4 ′) in FIG. 3 to the positions of the markers MK1 to MK4 (corresponding to F1 to F4) in FIG.

A matrix representing such a projection P1 is obtained as shown in the following expression (9) by solving simultaneous equations for the elements a ₁₁ , a ₁₂ , a ₁₃ ... A ₃₃ of the matrix.

Therefore, the conversion of the first stereoscopic method, which is a combination of the camera shooting (C2) and the post-processing (C3), is a matrix of the rotation Rx in the above equation (1) and a matrix of the projection Pz in the above equation (2). And the matrix of the projection P1 in the above equation (9), and can be represented as the following equation (10).

The above equation (10) can also be expressed by a conversion equation such as the following equation (11).

As described above, the first stereoscopic viewing method of FIG. 1 can be expressed by a mathematical expression such as the above expression (10) or the above expression (11).

9.2 Mathematical Analysis of the Second Stereoscopic Viewing Method The transformation of the second stereoscopic viewing method of FIG. 13 in which the object point is directly projected on the reference plane screen is expressed by the following equation (12) from FIG. be able to.

The above equation (12) can also be expressed by a conversion equation such as the following equation (13).

The conversion of the second stereoscopic viewing method expressed by the above formulas (12) and (13) includes the translation Ty of the object OB as shown in FIG. 25A (translation of -v cos α in the y direction), From the three transformations of the projection PZ of the object OB after translation as shown in FIG. 25B and the translation Ty (translation of vcos α in the y direction) of the object OB after projection as shown in FIG. It is made up.

9.3 Comparison of First and Second Stereoscopic Methods As described above, mathematically, the conversion of the first stereoscopic method is expressed as the following equation (14) or (15), and the second The conversion of the stereoscopic method is expressed by the following formula (16) or (17).

The difference between the above equations (14) and (16) is the term shown in J1 and the term shown in K1. In the above equations (15) and (17), this difference is the difference between the term indicated by J2 and the term indicated by K2.

  These differences are intuitively explained with reference to the drawings as follows. That is, as described above, the second stereoscopic viewing method is constituted by three conversions shown in FIGS. 25A and 25B and FIG. The first stereoscopic viewing system is different from the second stereoscopic viewing system in the amount of shift in the first parallel movement of FIG. That is, in the first stereoscopic viewing method, this shift amount is zcot α (see J1 and J2 in the above equations (14) and (15)). On the other hand, in the second stereoscopic viewing method, this shift amount is vcos α (see K1 and K2 in the above equations (16) and (17)).

As described above, in the second stereoscopic viewing method, the shift amount (vcos α) depends on the viewpoint (v) and the line-of-sight direction (α). On the other hand, in the first stereoscopic viewing method, the shift amount (zcot α) depends on the height (z) and the line-of-sight direction (α), and does not depend on the viewpoint (v) itself. As shown in FIG. 27, the shift amount (zcot α) in the first stereoscopic method is such that the perpendicular line drawn from the point (x, y, z) of the object intersects the reference plane (desk) screen. It is equal to the distance between N1 and the point N2 where the line extending from the point (x, y, z) of the object in the line-of-sight direction instead of the projection direction intersects the reference plane screen.

  As described above, in the first stereoscopic viewing system, the shift amount (zcot α) of the parallel movement in FIG. 25A depends on the height (z). Therefore, depending on the height (z) of the point (x, y, z) of the object, how the stereoscopic view looks in the first stereoscopic view method and how the stereoscopic view looks in the second stereoscopic view method In this respect, the first and second stereoscopic viewing methods are different.

  The present invention is not limited to that described in the above embodiment, and various modifications can be made.

  For example, terms (objects / subjects, polygons, etc.) cited as broad terms (objects, primitive surfaces, etc.) in the description in the specification or drawings are also broad terms in other descriptions in the specification or drawings. Can be replaced.

  Also, the method for creating (generating) the left-eye image, the right-eye image, and the stereoscopic image is not limited to that described in the present embodiment, and various modifications can be made.

  It is also possible to use the stereoscopic image created (generated) by the method of the present invention for uses other than the stereoscopic printed material and the game image.

  In addition, a case where a stereoscopic image is generated by a method equivalent to the first and second stereoscopic methods described in the present embodiment is also included in the scope of the present invention.

  In the invention according to the dependent claims of the present invention, a part of the constituent features of the dependent claims can be omitted. Moreover, the principal part of the invention according to one independent claim of the present invention can be made dependent on another independent claim.

  Further, the present invention can be applied to various games (such as fighting games, competitive games, shooting games, robot battle games, sports games, role playing games, etc.).

  The present invention is also applicable to various image generation systems (game systems) such as a business game system, a home game system, a large attraction system in which a large number of players participate, a simulator, a multimedia terminal, and a system board for generating game images. Applicable.

VPL left eye viewpoint position, VPR right eye viewpoint position,
OB object (object, subject), BS (BS1, BS2) reference plane,
RTG rectangle, MK1-MK4 mark, MK5-MK8 mark,
IL1 first left-eye image, IR1 first right-eye image,
IL2 second left eye image, IR2 second right eye image,
IL image for left eye, IR image for right eye,
100 processing unit, 110 game processing unit, 120 first image generation unit,
122 second image generation unit, 124 texture mapping unit 126 stereoscopic image generation unit, 130 sound generation unit, 160 operation unit,
170 storage unit, 172 main storage unit, 174 drawing buffer,
176 texture storage unit, 180 information storage medium, 190 display unit 192 sound output unit, 194 portable information storage device, 196 communication unit

Claims (4)

An image generation system for generating a stereoscopic image ,
Correction for eliminating a perspective in at least the depth direction of the image on the reference plane of the first left-eye image for stereoscopic viewing created by arranging an object on a reference plane that is oblique to the line-of-sight direction Processing is performed on the first left-eye image to create a second left-eye image, and the first right-eye image for stereoscopic viewing created by placing an object on the reference plane An image generation unit that performs correction processing for eliminating at least a perspective in the depth direction of the image on the reference plane with respect to the first right-eye image, and creates a second right-eye image ;
Including a stereoscopic image generating unit that creates a stereoscopic image based on the second left-eye image and the second right-eye image and outputs the stereoscopic image to a display unit including a lenticular lens. A featured image generation system . In claim 1,
The reference plane includes a first reference plane and a second reference plane forming a predetermined angle with respect to the first reference plane;
The image generator
The first correction process for the left eye for eliminating the perspective of the image on the first reference plane of the first left-eye image is performed on the region corresponding to the first reference plane of the first left-eye image. And applying the second correction process for the left eye for eliminating the perspective of the image on the second reference plane of the first left-eye image to the second reference plane of the first left-eye image The first correction process for the right eye for creating the second left-eye image and eliminating the perspective of the image on the first reference plane of the first right-eye image, A second correction process for the right eye for eliminating the perspective of the image on the second reference plane of the first right-eye image while applying to the region corresponding to the first reference plane of the right-eye image; Applying to the region corresponding to the second reference plane of the first right-eye image, the second right Image generation system characterized by creating a use images. An image generation system for generating a stereoscopic image ,
First left eye for stereoscopic vision created by photographing the subject and first to fourth marks that form a rectangle on a reference plane that is oblique to the line-of-sight direction from the left-eye viewpoint position A second left-eye image is created from the first left-eye image by performing correction processing for moving the first to fourth marks having perspective in at least the depth direction in the image to the vertex position of the rectangle, The first right eye for stereoscopic vision created by photographing the subject and the first to fourth marks that form a rectangle on the reference plane that is oblique to the line-of-sight direction from the right-eye viewpoint position The first right-eye image is changed to the second right-eye image by performing correction processing for moving the first to fourth marks having perspective in at least the depth direction in the image to the vertex position of the rectangle. An image generating unit that creates,
Including a stereoscopic image generating unit that creates a stereoscopic image based on the second left-eye image and the second right-eye image and outputs the stereoscopic image to a display unit including a lenticular lens. A featured image generation system . In claim 3,
The reference plane includes a first reference plane and a second reference plane forming a predetermined angle with respect to the first reference plane;
The first left-eye image is
By photographing the subject, the first to fourth marks that form a rectangle on the first reference plane, and the fifth to eighth marks that form a rectangle on the second reference plane from the viewpoint position for the left eye. The created image ,
The first right-eye image is
By photographing the subject, the first to fourth marks forming a rectangle on the first reference plane, and the fifth to eighth marks forming a rectangle on the second reference plane from the right eye viewpoint position. The created image ,
The image generator
The first correction process for the left eye is performed to move the first to fourth marks of the first left eye image to the vertex positions of the rectangle, and the fifth to eighth marks of the first left eye image are set. By performing the second correction process for the left eye that moves to the vertex position of the rectangle, a second left eye image is created from the first left eye image, and the first to fourth images of the first right eye image are created. The first correction process for the right eye that moves the mark to the vertex position of the rectangle is performed, and the fifth to eighth marks of the first right-eye image are moved to the vertex position of the rectangle. by performing the second correction process, an image generation system, characterized in that the first right-eye image to create a second right-eye image.