JPH0757118A - Image generating device - Google Patents

Abstract

PURPOSE:To enable an operator to correct the distortion of a generated image resulting from arrangement as the operator intends by generating the image while the operator interactively alters the correspondence relation between attribute information arranged on the surface of a body to represent the qualitativeness and the body surface. CONSTITUTION:The correspondence relation between shape information on the body stored in a body storage part 1 and the attribute information stored in an attribute storage part 2 is stored in the body storage part 1 and a correspondence relation calculation part 4 finds new correspondence relation on the basis of alteration information set by an alteration information setting part 3; and an image generation part 5 generates the image and also updates the correspondence relation stored in the body storage part 1.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for generating an image by arranging attribute information such as an image on the surface of an object shape defined in a three-dimensional space.

[0002]

2. Description of the Related Art A general technique used in computer graphics is a technique called texture mapping. This technique arranges an image (texture image) representing a pattern or texture on the surface of an object shape and generates an image representing the object. With this technique, an image expressing an object having a complicated pattern or texture can be generated using a simple shape. In addition, a mapping method in which a normal vector on the surface of an object is changed to express minute irregularities on the surface of the object is known as a bump mapping method.

Details of a technique for defining an image or normal line information as attribute information for an object surface and applying the defined attribute information to the object surface to generate an image are described in, for example, the literature [19].
1989, Alan Watt "Fundamentals of
Three Dimensional Computer Graphics ", Alan Watley
t, FUNDAMENTALS OF THREE-D
IMENSIONAL COMPUTER GRAPHI
CS, ADDISON WESLEY, 1989), 2
27-254].

Further, as a method for expanding such plane attribute information into three dimensions, three-dimensional attribute information (spatial attribute information) of minute structures and optical characteristics on the surface of an object is obtained.
There is a method of generating the image by attaching the defined spatial attribute information to the surface of the object. Using this method, it is possible to attach spatial attribute information that defines the structure and optical characteristics of the hair to the basic shape, without defining the shape of each hair individually, for complex shapes such as animal fur. You can generate images with. Further, this method has an advantage that by deforming the spatial attribute information on the shape surface of the object, it is possible to generate an image in which the surface shape is changed without changing the shape itself of the object. For details of the image generation method using the spatial attribute information, [1989, James T. Kaziya, Timothy El Kay, "Computer Graphics", Vol. 23, No. 3, AC M Siggraph (Ja
mes T. Kajiya, Timothy L .; Ka
y, COMPUTER GRAPHICS, ACM S
IGGRAPH, VOL. 23 NO. 3 1989),
271-280]. When actually creating an image by pasting attribute information using the texture mapping method,
First, the correspondence relationship between the object shape and the attribute information at the time of pasting is determined, and then the attribute information corresponding to the surface of the shape is arranged in accordance with the determined correspondence relationship to generate an image. There are the following methods for determining the correspondence between the object shape and the attribute information.

(1) A virtual surface such as a sphere or a cylinder is placed around the object shape, and attribute information is arranged on this surface.
Then, a method of projecting attribute information from the virtual surface onto the surface of the object (2) A method of setting a coordinate system on the surface of the partial shape of the object and associating it with the coordinate system in the attribute information.

[0006]

When an image is generated by the texture mapping method, the attribute information of the plane is expanded and contracted and attached to the surface of the solid. Therefore, the attribute information becomes sparse and dense at the time of attachment, resulting in distortion of the generated image. Will appear.

For example, when the correspondence is determined by the above methods (1) and (2), in the case of (1), there is a problem that the distortion becomes stronger as it gets closer to the peripheral edge of the attribute information to be pasted. Even in the case of (2), there is a problem that the more uneven the object shape is, the more dense the attribute information becomes when the information is pasted, and the stronger the distortion appears.

As described above, in the texture mapping method,
There is a problem that it is difficult to generate a high-quality image in which distortion is not noticeable except for a relatively simple shape.

One method for reducing the distortion in such a texture mapping method is disclosed in Japanese Unexamined Patent Publication No. 2-1143.
No. 84 publication. In this method, the attribute information is arranged so that the length of the attribute information along the coordinate axes of the coordinate system defined on the surface of the shape is constant. With this method, it is possible to arrange the natural attribute information with less expansion and contraction in the low curvature portion of the shape surface, but there is a problem that unnatural distortion occurs in the high curvature portion.

An object of the present invention is to provide an apparatus for generating an image while reducing distortion according to a user's intention by interactively setting a correspondence between attribute information and an object shape by the user. Especially.

[0011]

An image generating apparatus according to the present invention includes an attribute storage unit for inputting and storing attribute information for expressing the texture of the surface of an object defined in a three-dimensional space. An object storage unit for inputting and storing shape information that defines a shape and the correspondence relationship between the position of the surface of the object and the storage location of the attribute information in the attribute storage unit; and change information for changing the correspondence relationship. A change information setting unit for setting, a correspondence relationship calculating unit for changing the correspondence relationship based on the shape information and the change information to obtain a new correspondence relationship and updating the correspondence relationship stored in the object storage unit. , An image generating unit that generates an image by arranging the attribute information on the surface of the object based on the new correspondence relationship and the shape information.

[0012]

The function of the present invention will be described.

First, attribute information for expressing the texture of the surface of an object is set in the attribute storage unit. This attribute information is an image such as a picture or a pattern, or an array of the amount of normal vector variation that represents fine irregularities on the surface of an object.The attribute storage unit stores each element of the attribute information as an attribute value. Store based on index.

Next, shape information for defining the object shape is set in the object storage unit. The object shape is represented by a collection of partial shapes that form the surface of the object. The partial shape is a parametric function curved surface or a planar polygon such as a triangle.

Then, the index of the array of attribute values corresponding to the surface of the object is set as the correspondence of the attribute information to the object shape, and stored in the object storage unit.

The change information setting unit outputs the coordinate values input by the coordinate input device such as a mouse to the correspondence calculation unit as change information.

The correspondence calculation unit sets a change area to be changed for the current correspondence based on the change information. Then, similarly, the change amount of the correspondence relationship is set for the partial shape inside the change region based on the change information. Every time the change area and the change amount are designated, the correspondence calculation unit performs change calculation on the correspondence in the object storage unit to obtain a new correspondence. Then, the correspondence relationship of the object storage unit is updated to the new correspondence relationship.

The image generation unit generates an image in which the attribute information is attached to the surface of the object based on the obtained new correspondence, the shape information and the attribute information.

As described above, every time the change area and the change amount are set by the change information, a new correspondence is calculated and an image is generated, and as a result, the user evaluates the newly generated image. You can

Therefore, the user can rearrange the attribute information on the surface of the object as desired, and can freely correct the distortion.

[0021]

Embodiments of the present invention will now be described with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the present invention.

As shown in FIG. 1, the image generating apparatus of this embodiment comprises an object storing section 1, an attribute storing section 2, a change information setting section 3, a correspondence calculating section 4 and an image generating section 5.

The attribute storage unit 2 stores the attribute information 104. The attribute information 104 is image information when an image is generated by the texture mapping method, and is an array of brightness values given to each pixel. FIG. 3A is an explanatory diagram of attribute information used in texture mapping. Further, when an image is generated by the bump mapping method, the attribute information 104 is an array of variation values to the normal vector given to each pixel. Furthermore, when an image is generated with spatial attribute information that is three-dimensional attribute information, the attribute information 104 is a three-dimensional array (S, T) in which two-dimensional array attribute information (S, T) is stacked in the height direction U. T, U). FIG. 4A is an explanatory diagram of the spatial attribute information (S, T, U). Spatial attribute information (S,
Each of the array elements (T, U) is given an optical characteristic as an attribute value for expressing a minute structure or texture of the object surface.

FIG. 2 is a selection diagram showing the shape of an object stored in the object storage unit 1. As shown in FIG. 2, the surface shape of an object is defined by vertices v (i, j) arranged on a grid. An array of vertex coordinate values is stored in the object storage unit 1 as shape information 1
Store as 01. In the present embodiment, the quadrangle arranged on the lattice is described as a partial shape, but the partial shape may be an arbitrary planar polygon such as a triangle or a parametric function curved surface having the vertices of the polygon as control points. The present invention can be similarly implemented.

In FIG. 3B, the attribute information 3 is added to the object shape 301.
This is a description of a state in which 02 is attached. When two-dimensional attribute information such as the texture mapping method or the bump mapping method is used, the correspondence 102 is the array index in the attribute storage unit 2 of the attribute value corresponding to the vertex of the object surface, and is stored in the object storage unit 1. Store.

FIG. 4B is an explanatory diagram showing a state in which the space attribute information 402 is attached to the object surface 401. Figure 4
As shown in (b), the space attribute information 402 is the object surface 40.
Arrange so that the bottom is in contact with 1. FIG. 7 is an explanatory diagram of the correspondence relationship between the surface of the object shape and the spatial attribute information. As shown in FIG. 7, at the vertex v (i, j) on the object surface 901, the array element (Sfij, Tf) of the bottom of the space attribute information 902 is formed.
ij, 0) and the array element (Scij, Tci) on the upper surface.
j, U) correspond to two array elements. The correspondence between the apex and the two array elements is called the bottom correspondence and the top correspondence, respectively. These correspondences are defined as array indexes in the attribute storage unit 2 of the attribute values corresponding to the vertices. Then, in order to express the spatial correspondence between the vertex v (i, j) and the array element (Scij, Tcij, U) on the upper surface, the position vector (Xij, Yij) having the vertex v (i, j) as a starting point is represented. , Zij) is defined. The end points of the defined position vector (Xij, Yij, Zij) correspond to the array elements (Scij, Tcij,
U) corresponds. Position vector (Xij, Yij, Zi
j) is defined by a coordinate system that defines the shape of the object, and is called a spatial correspondence as a correspondence representing the spatial positional relationship between the vertex and the upper surface. When the spatial attribute information is used, the top surface correspondence, the bottom correspondence, and the space correspondence corresponding to each vertex of the object surface are stored in the object storage unit 1 as the correspondence 102.

The change information setting unit 3 inputs the coordinate value of the pointer in the device coordinate system with a coordinate input device such as a mouse, a tablet, or a three-dimensional digitizer, or a keyboard, and outputs it as change information 103 to the correspondence calculation unit 4. .

The correspondence calculation unit 4 obtains the change area and the change amount based on the change information 103 output by the change information setting unit 3. The change area is an area whose correspondence is to be changed on the surface of the object, and the change amount is the variation of the correspondence inside the change area. The correspondence calculating unit 4 obtains the change area and the change amount, reads the correspondence 102 from the object storage unit 1, and calculates a new correspondence.

In the following, first, a method for obtaining the changed area and the changed amount by the attribute information defined two-dimensionally such as an image or a normal vector will be described, and then a method for obtaining the changed area and the changed amount by the spatial attribute information. explain.

As a method for obtaining the change area and the change amount for the two-dimensional attribute information, (1) a method for obtaining the change area and the change amount independently, and (2) obtaining the source area and the destination area of the attribute information. Although a method of obtaining the changed area and the changed amount will be described with reference to FIG.

First, the method (1) of independently obtaining the change area and the change amount will be described. FIG. 5A is an explanatory diagram of a change area designated on the object surface. Change area 50
2 is obtained as a set of vertices of the object surface 501.

The change area 502 can be obtained by repeatedly performing an operation known as a pick on the basis of the change information 103 input by the change information setting unit 3 using a coordinate input device such as a mouse or a keyboard. In this operation, the target object shape is displayed as an image, the pointer that the user operates by operating the coordinate input device is superimposed on the image, and the partial shape or vertex displayed closest to the pointer is displayed. The search is performed and the result is used as a candidate for the selection result of the user. This operation is performed on the object surface 50
A set of vertices or partial shapes repeatedly specified for one vertex is obtained as a change area. For details of this operation, refer to [1990, Foley, Van Damme, Finner, Hues, "Computer Graphics Principles and Practice.
Second Edition ", Addison Wesley (Foley, van Dam, Feiner, Hug
hes, COMPUTER GRAPHICS PRI
NCIPLES AND PRACTICE SECO
ND EDITION, ADDISON WESLE
Y, 1990), pages 328-331, 338-339.
Page].

The change area 502 is the object surface 501.
It can also be obtained by logically operating a product set of a sphere, a rectangular parallelepiped, or a closed surface having a free shape set in advance. In this case, a set of vertices or partial shapes on the object surface 501 included in the intersection of two shapes is obtained as the change area.

An arbitrary amount of change can be given to each of the vertices inside the change area 502 thus obtained by the change information 103 similarly input by the change information setting unit 3. In FIG. 5B, the vertices (X,
When the correspondence in Y) before the change is (S, T), the vector which makes the correspondence after the change (S ', T') is the change amount (dS, dT) _XY = (S'- S, T'-
T). This change amount (dS, dT) _XY is
The vertices inside the change region 503 may be arbitrarily given. For example, the vertices at the center of the change region 503 may be large, and the vertices at the periphery may be small.

Next, the method (2) of obtaining the change area and the change amount by obtaining the move source area and the move destination area of the attribute information will be described. FIG. 6 is an explanatory diagram showing a movement source region 601 and a movement destination region 602 on the object surface 603. The source area 601 and the destination area 602 are (1)
It is calculated by the same procedure as the setting of the change area in the method. The point (X, Y) and the vertex (X ′, Y ′) correspond to the attribute values (S, T) and (S ′, T ′) of the attribute information 604, respectively.

First, a method of obtaining the change amount (dS, dT) _{X ′ Y ′} of the vertex (X ′, Y ′) inside the movement destination area 602 will be described. Change amount (dS, dT)
_In calculating X'Y ', the point (X, Y) of the source area 601 corresponding to the vertex (X', Y ') of the destination area 602 is obtained, and the correspondence relationship of the points (X, Y) ( S, T) is calculated. The point (X, Y) is obtained by performing interpolation between the two areas of the movement source and the movement destination. Interpolation between two regions is described, for example, in the literature [1992, Thomas W. Sederberg, Eugene Greenwood,
"Computer Graphics", Vol. 26, No. 2,
AMC M Siggraph (Thomas W. Sed
erberg, Eugene Greenwood, C
OMPUTER GRAPHICS, ACM SIGG
RAPH, VOL. 26 NO. 2 1992), 25
-Page 34]. The point (X, Y) obtained by interpolation between the two areas generally does not coincide with the position of the apex of the source area 601. Therefore, the correspondence relationship (S, T) at the point (X, Y) is obtained by interpolating the correspondence relationship between the vertices of the movement source region 601 near the point (X, Y). After the point (X, Y) and the correspondence (S, T) are obtained in this way, the change amount (dS, dT) of the point (X ', Y') inside the movement destination area 602 is calculated.
_{X'Y '} is calculated by the following formula.

(DS, dT) X'Y _' = (S-S', T-T ') Next, a change area and a change amount around the movement source area and the movement destination area will be described. FIG. 8A is an explanatory diagram showing a state before the attribute information is moved. Source area 1001
An area between the destination area 1002 and the destination area 1002 is an intermediate area 1003. A region surrounding the movement source region 1001 and the intermediate region 1003 is referred to as a peripheral region 1004. FIG. 8B is an explanatory diagram showing the state of the attribute information after the attribute information is moved and made to correspond. As shown in FIG. 8B, FIG.
The attribute information of the movement destination area 1002 and the intermediate area 1003 corresponds to the peripheral area 1005 outside the movement source area 1001 so as to be compressed. The attribute information of the peripheral area 1004 in FIG. 8A is the same as the source area 1001, the intermediate area 1003, and the peripheral area 100 in FIG. 8B.
The region corresponding to the union of 4 is expanded.

To summarize the above-described correspondence, in the area around the movement source area and the movement destination area, [movement destination area 1002 + intermediate area 1003 → peripheral area 1
[005] [Peripheral area 1004 → peripheral area 1004 + source area 1
001 + intermediate area 1003] is changed. In each case, the internal correspondence of each area to which the attribute information is moved is obtained by the above-described method of calculating the correspondence by interpolation between the two areas.

Next, the calculation of the change area and the change amount in the case of the spatial attribute information will be described. The correspondence calculation unit 4 obtains the change area and the change amount for the bottom correspondence, the top correspondence, and the space correspondence, respectively.

In FIG. 7, the vertex v (i,
Of the spatial attribute information corresponding to j), the bottom correspondence (S
The change area and the change amount regarding fij, Tfij, 0) can be calculated in the same manner as in the case of the two-dimensional attribute information described above. In addition, the upper surface correspondence (Scij, Tc
ij, U) can be similarly obtained.

Spatial correspondence (Xij, Yij, Zij)
Can be obtained by setting an initial value in advance and changing the initial value. For example, vertex v
Position vector (Xij, Yij, which stands on (i, j)
Zij) is displayed as an image, and based on the change information 103, the destination vector end point (X′ij, Y′ij, Z′ij)
Ask for. Then, the calculated destination vector end point (X ′
From ij, Y'ij, Z'ij), the change amount of spatial correspondence (dXij, dYij, dZij) = (X'ij-X)
ij, Y'ij-Yij, Z'ij-Zij).

After the change area and the change amount are obtained from the change information 103 as described above, the correspondence calculation unit 4 reads the correspondence stored in the object storage unit 1 and calculates a new correspondence. . The new correspondence is calculated by adding the change amount to the vertex inside the change area with respect to the correspondence read from the object storage unit 1. For the vertices not specified in the change area, the correspondence relationship is used without change.

The calculated new correspondence is the latest correspondence 1.
The result is output to the object storage unit 1 as 05, and the correspondence relationship stored inside is updated. In addition, it outputs to the image generation unit 5.

The image generator 5 generates an image representing the texture by sticking the attribute information on the surface of the object based on the latest correspondence 105, the shape information 101 and the attribute information 104. For more information on texture mapping and bump mapping image generation, see [1990, Foley, Van Damme, Finner, Hughes, "Computer
Graphics Principles and Practice Second Edition ", Addison Wesley (Foley, van Dam, Feiner, Hu
ghes, COMPUTER GRAPHICS PR
INCIPLES AND PRACTICE SEC
OND EDITION, ADDison WESLE
Y, 1990), pp. 741-745] and literature [19.
1989, Alan Watt, "Fundamentals of Three Dimensional Computer Graphics", Addon Wesley.
t, FUNDAMENTALS OF THREE-D
IMENSIONAL COMPUTER GRAPH
ICS, ADDISON WESLEY, 1989),
Pp. 227-254]. The details of the image generation method using the spatial attribute information are described in [1989, James T. Kaziya, Timothy L. Kay, "Computer Graphics", Vol. 23, No. 3, James T. Siggraph. Kajiya, Ti
mothy L. Kay, COMPUTER GRAP
HICS, ACM SIGGRAPH, VOL. 23
NO. 3 1989), pages 271-280].

[0045]

As described above, the image generating apparatus according to the present invention can interactively change the correspondence between the surface shape of an object and the attribute information based on the change information set by the user. This has an effect that attribute information can be arranged on the object surface according to the user's intention, and an image can be generated while reducing distortion according to the user's intention.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is an explanatory diagram of shape information according to the embodiment of this invention.

FIG. 3 is an explanatory diagram of two-dimensional attribute information according to the embodiment of the present invention and an explanatory diagram of a relationship between the two-dimensional attribute information and a shape.

FIG. 4 is a selection diagram of spatial attribute information and an explanatory diagram of a relationship between spatial attribute information and a shape according to the embodiment of the present invention.

FIG. 5 is an explanatory diagram of a changed area and an explanatory diagram of calculation of a correspondence relationship according to the embodiment of this invention.

FIG. 6 is an explanatory diagram of a movement source area and a movement destination area according to the embodiment of this invention.

FIG. 7 is an explanatory diagram of a relationship between an object surface and space attribute information according to the embodiment of this invention.

FIG. 8 is an explanatory diagram of a correspondence changing operation according to the embodiment of this invention.

[Explanation of symbols]

 1 Object Storage Section 2 Attribute Storage Section 3 Change Information Setting Section 4 Correspondence Calculation Section 5 Image Generation Section 101 Shape Information 102 Correspondence Relationship 103 Change Information 104 Attribute Information 105 Latest Correspondence 301 Object Shape 302 Attribute Information 401 Object Surface 402 Spatial Attribute Information 501 Object surface 503 Change area 601 Source area 602 Destination area 603 Object surface 604 Attribute information 901 Object surface 902 Spatial attribute information 1001 Source area 1002 Destination area 1003 Intermediate area 1004 Edge area 1005 Edge area

Claims (1)

[Claims] 1. An attribute storage unit for inputting and storing attribute information for expressing the texture of the surface of an object defined in a three-dimensional space, shape information defining the shape of the object, and the position of the surface of the object. And an object storage unit for inputting and storing a correspondence relation between the attribute information and a storage location in the attribute storage unit, a change information setting unit for setting change information for changing the correspondence relation, and the shape information. On the basis of the new correspondence and the shape information, a correspondence calculation unit that changes the correspondence based on the change information to obtain a new correspondence and updates the correspondence stored in the object storage unit. And an image generating unit that generates the image by arranging the attribute information on the surface of the object.