JP2001067493A - Picture display device by light beam tracking - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce cross judgment with subject and to display a picture at a high speed by executing a cross judgment on a voxcel only with a subject in a bounding volume crossing the voxcel. SOLUTION: An input part 11 generates a scene data base from inputted data. A pre-processing part 12 obtains the maximum value and the minimum value of three-dimensional coordinates on a subject based on the scene data base, calculates a maximum rectangular area containing all subjects on a scene, equally divides the maximum rectangular area and forms a voxcel. A bounding volume wrapping an object is generated in parallel to voxcel division. Cross judgment on the generated voxcel and the generated bounding volume is executed. Cross judgment on the object in the bounding volume crossing the voxcel and the voxcel is executed. The result is stored as the intra-voxcel subject data base of the voxcel.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a ray tracing (ray tracing) image display apparatus for displaying an image by judging the intersection of a light beam and a subject. In particular, it is an object of the present invention to provide an image display device capable of displaying an image at a high speed by reducing intersection determination with a subject.

[0002]

2. Description of the Related Art The ray tracing method is a technique in which a light ray is skipped from a viewpoint to each pixel of a screen, an intersection between the light ray and the object is determined, and an object which intersects at a point closest to the screen is determined as an object to be drawn. This is a method of calculating luminance and creating an image as the color of the pixel. By using the ray tracing method, complicated expressions such as reflection and refraction of an object can be accurately performed.

However, in the ray tracing method, when determining a subject that intersects at a point closest to the screen, it is necessary to calculate intersections with all subjects for each ray, which requires a very long processing time. , High-speed technology is being studied.

As a conventional high-speed technology, when one object is formed by a plurality of subjects, the object is enclosed by a three-dimensional rectangle called a bounding volume or an object that can be represented by a simple formula such as a sphere, and a light beam is included. A method is known in which the number of intersection determinations is reduced by performing an intersection determination between the object and the bounding volume, and determining an intersection between only the object within the intersecting bounding volume and the light beam to obtain a subject to be drawn. Further, as another method, a space division method using voxels obtained by equally dividing a three-dimensional rectangular area including all subjects is known.
By performing the space division, it is only necessary to perform the intersection determination with only the subject in the voxel through which the light beam passes, and the processing by the ray tracing method can be speeded up.

Hereinafter, the space division method will be described with reference to FIGS. In the space division method, scene data including information on three-dimensional subject data, a camera, a light source, and the like is input to the input unit 11 shown in FIG. 2 to create a scene database (FIG. 3).
Step 211). After that, an image is generated through the preprocessing unit 12, the image generation unit 13, and the drawing unit 14.

The pre-processing unit 12 first generates a three-dimensional rectangular area (maximum rectangle) encompassing all objects (step 221 shown in FIG. 3). The generated rectangular area is equally divided based on the designated number of divisions or the voxel size to form a voxel (step 222). FIG. 5 described in detail in the embodiment.
In (simply shown in two dimensions), 51 is a maximum rectangular area, and 54 is one divided voxel. For each of the divided voxels, the intersection between the voxel and all the subjects is determined, and the subject inside the voxel is checked.
Create an in-voxel subject database that stores which voxel contains which subject (step 22)
3). In the case of a subject extending over a plurality of voxels, each part of the subject included in one voxel is not specified by the in-voxel subject database, but the voxel determines which subject of all the subjects is included in the voxel. It is specified by the inner subject database.

Next, the image generation unit 13 calculates a light ray to be skipped from the viewpoint to a pixel on the screen to virtually generate a light ray (step 231), and determines a voxel through which the light ray passes by 3DDD.
It is obtained by the method A (a method of calculating the next voxel at high speed) (step 232), and the database of the voxels through which the light beam passes is examined to find the intersection of the light beam with the object in the voxel (step 233). Determine whether the intersection is within the voxel (step
234) If not, return to step 232 to find the next voxel and do the same. If the intersection is within the voxel, the brightness of the intersection is calculated (step 235), and the process moves to the next pixel.
This is performed for all the pixels (the end determination is made in step 23).
6), generate an image. Drawing unit 14 for generated image
(Step 241).

[0008]

In the conventional space division method, even in a voxel in which no light ray passes as a result, the preprocessing unit 12 determines the intersection between the voxel and all the objects in order to create an in-voxel object database. It is carried out. Therefore, the conventional space division method has a problem in that, when the number of voxel divisions is large or the number of subjects is large, useless processing in the preprocessing unit increases, and the processing time is shortened. The conventional method using the bounding volume also has a problem that the effect of speeding up is small when one object includes a large number of subjects. The present invention reduces intersection determination with a subject,
An object is to provide an image display device capable of displaying an image at high speed.

[0009]

Therefore, in order to solve the above-mentioned problems, the present invention provides a method that includes all objects on a scene.
In an image display device that divides a dimensional rectangle into voxels to create a voxel subject database and displays an image by ray tracing, a bounding volume generating unit that generates a bounding volume including an object constituted by the subject, and each voxel Is determined within the one voxel by performing an intersection determination between the subject and the one voxel in the bounding volume determined to intersect with one voxel. It is another object of the present invention to provide an image display apparatus by ray tracing, comprising: an in-voxel object database creating means for creating an in-voxel object database which is a database for specifying an object.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the present invention, in a preprocessing unit for space division, first, an intersection between a bounding volume and a voxel is determined. Based on the determination result, only a subject in the bounding volume that intersects the voxel is determined as a voxel. To create an in-voxel object database. As a result, the present invention can significantly reduce the number of times of intersection determination between a voxel and a subject, and can greatly increase the speed of image generation and display as compared with the related art.

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a flowchart of one embodiment. The outline of the entire configuration is the same as that of the conventional example shown in FIG. The input unit
11, the image generating unit 13 and the drawing unit 14 have the same configuration and operation as those of the conventional example shown in FIG.
3. The description of the operation of the drawing unit 14 is omitted. The following description focuses on the operation of the pre-processing unit 12, which is different from the conventional example.

When drawing a three-dimensional scene as shown in FIG. 4, a viewpoint 41, a screen 42, and subject information are input to the input unit 11. Here, the subject information of this scene includes a subject 43 which is a rectangular parallelepiped constituted by six polygons,
It is assumed that there are two pieces of object information of the object A including the object 44 and the object B including the subject 45 composed of five polygons. The input unit 11 creates a scene database from the input data (step 31 shown in FIG. 1).
1) The processing is moved to the preprocessing unit 12.

The preprocessing unit 12 first obtains the maximum and minimum values of the three-dimensional coordinates of the subject based on the scene database, and calculates the maximum rectangular area including all the subjects on the scene (step 312: maximum rectangular area). (Calculating means), the largest rectangular area is equally divided to form a voxel (step 313:
Voxel dividing means). The dividing method includes a method of dividing the image into a predetermined number and a method of adaptively dividing the image according to the number of subjects, but any method may be used.

In addition, bounding volumes 46 and 47 enclosing the objects A and B are created in parallel with the voxel division (step 314: bounding volume generating means). In the example shown in FIG. 4, the object is wrapped using the rectangular bounding volumes 46 and 47, but a sphere or other objects may be used.

FIG. 5 shows a simplified two-dimensional state in which the processing of steps 313 and 314 is completed. 51 is the maximum rectangular area calculated in step 312, and 46 is the object A
, 47 is the bounding volume of object B, 54 is one voxel divided,
55 and 56 indicate polygons in the object A (surfaces of the subjects 43 and 44).

Next, the intersection of the voxel generated in step 313 and the bounding volume generated in step 314 is determined (step 315: means for creating an in-voxel object database), and if there is a bounding volume that intersects the voxel, Then, the intersection of the subject and the voxel within the bounding volume is determined (step 316: means for creating a subject database in the voxel). The result is stored as an in-voxel subject database of the voxel (step 317: means for creating an in-voxel subject database).

Taking the voxel 54 shown in a simplified two-dimensional manner in FIG. 5 as an example, first, the intersection of the voxel 54 and the bounding volumes 46 and 47 of the objects A and B is determined. Object A bounding volume 46 and voxel 5
4 intersects, followed by bounding volume 46
Then, the intersection of the voxel 54 with the subject (here, a polygon) included in the object A is determined. As a result of the intersection determination, it is found that the polygons 55 and 56 intersect with the voxel 54, and thus information on the two polygons is stored in the voxel object database in the voxel 54.

The operations in steps 315 to 317 are performed for all voxels, and a voxel object database is created (the operation in steps 315 to 317 is a voxel object database creation means). Thereafter, the image generation unit 13 generates an image based on the in-voxel subject database in the same manner as in the related art, and the drawing unit 14 performs drawing in the same manner as in the related art.

In this embodiment, the pre-processing unit 12 differs from the prior art.
In the above, instead of performing the intersection determination between the voxel and all the subjects to create the in-voxel object database, the subject for which the intersection determination with the voxel is performed is only the subject in the bounding volume that intersects the voxel. Therefore, in the present embodiment, the number of times of determining the intersection between the voxel and the subject for creating the in-voxel subject database in the space division method can be greatly reduced, and the speed of image generation and display can be greatly increased as compared with the conventional one. I can do it.

[0020]

As described above, the image display device according to the present invention comprises:
The number of times of determination of intersection between a voxel and a subject for creating a voxel subject database can be greatly reduced, and image generation and display can be performed at a significantly higher speed than the conventional one.

[Brief description of the drawings]

FIG. 1 is an operation flowchart of an embodiment.

FIG. 2 is a diagram showing a configuration of a conventional example.

FIG. 3 is an operation flowchart of a conventional example.

FIG. 4 is a three-dimensional scene diagram of one embodiment.

FIG. 5 is a voxel division diagram of one embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Input part 12 Preprocessing part 13 Image generation part 14 Drawing part

Claims (1)

[Claims] 1. An image display apparatus which divides a three-dimensional rectangle including all subjects in a scene into voxels to create a voxel subject database and displays an image by ray tracing, wherein an object constituted by the subjects is included. A bounding volume generating means for generating a bounding volume to perform, an intersection determination between each voxel and the bounding volume, and then an intersection between the subject in the bounding volume determined to intersect with one voxel and the one voxel An image display device based on ray tracing, comprising: an in-voxel object database creating means for creating an in-voxel object database which is a database for identifying an object present in the one voxel by performing the determination. .