JPH0368080A - Moving picture generating device - Google Patents

Abstract

PURPOSE:To generate a moving picture in real time by detecting movement of a visual point. CONSTITUTION:Random signals a1 to a3 generated by pulse generating means 201 to 203 pass space filters 205 to 207 to obtain unit time pictures, and they are supplied to buffer means 208 to 210. Unit time pictures are outputted to and successively stored in buffer means 208 to 210 by each signal b1 to b3 generated from pulse generating means 201 to 203. A visual point information extracting means 204 outputs the visual point position of an observer to effective picture extracting means 211 to 213, and effective picture outputs are outputted to picture adjusting means 214 to 216. Thus, the picture viewed from the visual point position is obtained, and the value in the height direction is converted in accordance with the distance from the visual point position with respect to this picture. Outputs of picture adjusting means 214 to 216 are inputted to a picture synthesizing means and are subjected to shade face processing to generate the picture of a natural object viewed from the visual point.

Description

[Detailed description of the invention] [Industrial application fields] The present invention relates to an image generation device, and particularly to a moving image reproduction device.

[Prior art] Conventionally, this type of technology includes, for example, "NIKKEI COMPUT
ERGRAPHIC9) (1988-9) P, 1
There is one disclosed in No. 32-140, which expresses natural scenery and natural phenomena.

FIG. 2 is a block diagram showing the configuration of a conventional image generation device. Each cross-sectional wave generating means lot~(04 generates a one-dimensional cross-sectional wave expressed by the establishment process as a power spectrum of the shape of a natural object. These cross-sectional waves are generated by a coordinate rotation means 105 whose traveling direction is defined by a different vector.
The coordinates are rotated by ~toll, and cross-sectional waves are output in each direction. Multiplying means 109 and 110 perform a set operation on these outputs to construct a surface for each object.

Adding means 111 synthesizes the surfaces of each object output from these multiplying means 109 and 110 to form a three-dimensional image. Then, the image output means 112 converts this three-dimensional image into C
Output to an RT display device, printer device, etc.

[Problems to be Solved by the Invention] By the way, the above-mentioned conventional technology is based on the generation of linear cross-sectional waves, and in order to express the shape of a natural object through operations such as addition and multiplication, -dimensional cross-sectional waves are generated. Since the computations of the stochastic process, addition, and multiplication are performed for each point, there is a problem that the computations are complex and enormous.

Further, the above-mentioned conventional technology has a problem in that it is difficult to generate an image based on viewpoint information, and it is difficult to generate an image by moving the viewpoint.

This invention eliminates the problem of difficulty in dynamic image generation, and also makes it possible to generate moving images in real time while detecting the movement of the viewpoint and generating the movement of natural objects. The purpose is to provide equipment.

[Means for Solving the Problems] A moving image generating device according to the present invention includes a pulse generating means that generates a random signal and a unit time pulse, and a pulse generator that inputs the random signal and the unit time pulse and generates a unit time pulse for each unit time pulse. A spatial filter means for outputting a three-dimensional image based on a power spectrum on a two-dimensional plane, a buffer means for buffering the three-dimensional image output per unit time from the spatial filter, and a viewpoint information extraction means for inputting a viewpoint position. effective image extracting means for extracting a valid image from the three-dimensional images already buffered in the buffer means according to the viewpoint position, and converting the three-dimensional image outputted from the effective image extraction means into an image seen from the viewpoint position. It has a plurality of image adjustment means for adjusting, and an image composition means for combining adjusted three-dimensional images output from each image adjustment means.

[Function] In the present invention, a three-dimensional image is generated for each single-time pulse using a power spectrum on a two-dimensional plane, and is buffered in a buffer means. An effective image is extracted from the buffer means according to the viewpoint position, the effective image is adjusted to an image seen from the viewpoint position, and the adjusted three-dimensional image is synthesized.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the configuration of a moving image generation device according to an embodiment of the present invention. Pulse generating means 201-2
03 are connected to spatial filter means 205-207, respectively, and each spatial filter means 205-207 is connected to buffer means 208-210, respectively.

The output of each buffer means 208 to 210 is connected to effective image extraction means 211 to 213, and the viewpoint information extraction means 204
The outputs thereof are connected to effective image extraction means 211-213 and image adjustment means 214-21B. The output images of each effective image extraction means 211 to 213 are each image adjustment means 214 to
21B respectively. Each image adjustment means 214~
The output of 21B is connected to image composition means 217, and the output of image composition means 217 is connected to image output means 218.

First, the pulse generating means 201 to 203 generate a random signal a that randomly generates logic "0" and logic "1" signals.
Generate l-a3. The generated random signals are input to spatial filter means 205-207.

The spatial filter means 205 to 207 use first-order Butterworth filters expressed by the following equations.

Here, σ is the height variance value, Xo and yo are the correlation lengths in the X direction and the y direction, and these constants are unique to each natural object.

Conceptually, the input random pulse train is input into the first term of the first equation, calculates the power spectrum in the X direction, and multiplies the second term of the first equation by the power spectrum in the create an image. Here, this image is called a unit time image.

Each of the spatial filters 205 to 207 has a number corresponding to the number of pixels in the X-axis direction, for example, one first-order digital filter that inputs the Langmuth pulse signal at-a3 and a large number of cascaded delay memory elements. It is possible to create an output, and use one primary digital filter provided corresponding to each output to create the number of outputs corresponding to the number of pixels in the y-axis direction.

X output from the spatial filter means 205-207.

Images on the y-plane are input to buffer means 208-210. This creates an image of the natural object at time tl.

- The pulse generation means 201 to 202 pass the signal bl-b3 of one pulse per unit time to the spatial filter means 205 to 20.
Output to 7. The spatial filter means 205 to 2Q7 output the unit time image to the buffer means 208 to 210 for each single crying time pulse b1 to b3.

FIG. 3 is an explanatory diagram showing the structure of the buffer means 208-210. As shown in the figure, buffer means 208-210 store unit time images output from spatial filter means 205-207, for example, an 11-hour image (31>, t1+Δ
are sequentially stored as time images (32).

The viewpoint information extraction means 204 inputs the viewpoint position of the observer using an input device (not shown). For example, a mouse is used as this input device. The viewpoint position of the viewpoint information extraction means 204 is output to the effective image extraction means 211 to 213.

FIG. 4 is an explanatory diagram of the operation of the effective image extracting means 211-213. As shown in FIG. 3, effective image extracting means 211-213 extract effective screen areas such as screens and printers from the viewpoint positions from buffer means 208-210. Specifically, the plane surrounded by (PL, P2. P3. P4) in FIG.
When the viewpoint position has moved from 4, the effective image at that viewpoint position is cut out as an effective image 2 (42).

The effective image output per unit time of each effective image extracting means 211-213 is output to image adjusting means 214-216. The input effective image is coordinate-transformed into an image viewed from the viewpoint position. At the same time, the value of Φ(x, y) in the height direction is converted depending on the distance from the viewpoint position. Specifically, the height of a point close to the viewpoint is converted to be large, and the height of a position far from the viewpoint is converted to be small, and perspective conversion is performed to an image seen from the viewpoint position.

The output images of the image adjustment means 214 to 216 are input to the image composition means 217. The image synthesizing means 217 performs hidden surface processing on the images manually generated by the respective image adjusting means 214 to 216 and synthesizes them. Here, an image of the natural object as seen from the viewpoint is generated.

In the above image generation, the buffer means 208 to 21
0, the pulse generating means 201 to 203
The image generation unit per unit time by the spatial filter means 205 to 207, the effective image extraction means 211 to 213, the image adjustment unit 2 (4 to 216) and the image adjustment unit by the viewpoint information extraction f stage 204 can operate in parallel. Therefore, high-speed 3D moving images can be generated.

FIG. 5 is an explanatory diagram of another example of the configuration of the buffer means 208-210. In this embodiment, the edges of the image plane formed on the plane shown in FIGS. 3 and 4 are connected to form a columnar shape to constitute a buffer means. Then, the unit time image (31
), (32) are filed toward the center over time,
When extracting the effective images (41) and (42), they are basically processed and extracted in the same manner as the example shown in FIG.

Note that the configuration of the buffer means 208 to 210 is not limited to the above-described embodiment, and the above-described configuration may be appropriately modified and used, and for example, they may be configured in a spherical shape.

[Effects of the Invention] As explained above, according to the present invention, a three-dimensional image is generated and buffered for each single-time pulse using the power spectrum on a two-dimensional plane, and three-dimensional images buffered according to the viewpoint position are By extracting an effective image from a dimensional image and adjusting the effective image to the image seen from the viewpoint position to synthesize a 3D image, up until now only static images could be generated, but the effective image can be generated by moving the viewpoint. By extracting images, not only moving images of moving natural objects, but also the movement of the viewpoint and natural movements can be shown to the observer in real time and naturally.

[Brief explanation of drawings]

Claims (1)

[Claims] Pulse generating means for generating a random signal and a unit time pulse; and a space for inputting the random signal and a unit time pulse and outputting a three-dimensional image based on a power spectrum on a two-dimensional plane for each unit time pulse. a filter means; a buffer means for buffering a three-dimensional image output per unit time from the spatial filter means; and a buffer means for buffering a three-dimensional image output per unit time from the spatial filter means, based on a viewpoint position from a viewpoint information extraction means inputting a viewpoint position. effective image extracting means for extracting an effective image from among the three-dimensional images that have been displayed; image adjusting means for adjusting the three-dimensional image output from the effective image extracting means to an image seen from a viewpoint position; and the image adjusting means. 1. A moving image generating device comprising: image synthesizing means for synthesizing adjusted three-dimensional images output from the means.