JPH0636025A - Defocusing processor - Google Patents

Abstract

PURPOSE:To easily express a perspective feeling of an optional extent in a short time as to three-dimensional computer graphics production. CONSTITUTION:A shape input device 10 outputs a body shape, etc., to a position information specifying device 12, which specifies the position of the body, a light position, a camera position, a focus position, etc., according to the input information and outputs the specification information to an image generating device 14; and the image generating device 14 generates a two-dimensional projection image. Then, the defocusing processor 16 calculates the difference in depth between the focus and body by a depth calculation part 18 from the depth information from the position information specifying device 12 and a defocusing process part 22 filters the two-dimensional projection image, inputted from the image generating device 14, as many times as specified by a process frequency input device 20 by using a filter generated on the basis of the difference in depth to generate a processed image, which is stored in a storage device 24.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention, in a defocus processing apparatus, and particularly in the production of three-dimensional computer graphics, imparts a perspective to an original image projected on a two-dimensional screen in a short time and easily. Defocus processing device capable of performing

[0002]

2. Description of the Related Art Three-dimensional computer graphics (CG), for example, CG calendar, CG hologram,
In the production of CG animation, CG commercials, CG posters, high-definition CG, etc., a two-dimensional image produced by a rendering process is produced in a state in which all the pixels constituting the image are in focus. The object drawn gives a flat impression without perspective.

Conventionally, a dispersed ray tracing method is known as a method of giving a perspective feeling to an image produced by three-dimensional CG, such as when a photograph is taken. This distributed ray tracing method is a method of expressing perspective by giving a blur to an image portion that is out of focus. For this distributed ray tracing method, for example, ray tracing CG
Introductory (Science) first edition issued June 25, 1960,
P. by Norimori Chiba. 129.

[0004]

However, in the above distributed ray tracing method, in order to set the degree of defocus,
The physical law is virtually applied to each pixel to calculate, and more than one ray has to be calculated to obtain the blur for one pixel, so it is enormous to obtain a perspective image. It takes a long time to calculate.

Further, if the image obtained by applying the above-mentioned distributed ray tracing method has too weak a degree of blurring, or conversely, it is necessary to recalculate from the beginning, There is a problem that it is not easy to freely change the degree of defocus.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a defocus processing apparatus capable of easily setting a certain degree of blurring in a short time and producing a sense of perspective in an image in the production of three-dimensional computer graphics. And

[0007]

SUMMARY OF THE INVENTION The present invention is a defocus processing device for blurring an image based on distance information from the focus position of an object in the production of three-dimensional computer graphics, which is produced by rendering processing. With respect to the obtained two-dimensional original image, a function of calculating a depth difference between the focus position and the object based on depth information between the camera and the focus position, and a focus position of each pixel forming the object and a depth of the object. The object is achieved by a configuration including a function of performing a filtering process on the original image based on the difference.

[0008]

In the present invention, first, a two-dimensional original image is created in advance by a rendering process such as a ray tracing method, and the created original image is filtered based on the distance from the focus to the object position. The object on the focus position is in focus, and it is possible to create an image with a desired degree of defocus by making the object farther from the focus position out of focus, so an image that is out of focus in a short time, That is, an image with a perspective can be produced.

Further, since the above-mentioned defocus can be set by post-processing for the original image already created, the degree of defocus of the image can be easily changed.

[0010]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the outline of the configuration of a defocus processing apparatus according to an embodiment of the present invention.

The defocus processing apparatus according to the present embodiment is a shape input device 10 and an object position / light / light position specifying information such as each position of an object, a light, a camera, and a focus based on an input from the shape input device 10. A camera / focus position information designating device 12 and a two-dimensional projection image creating device 14 for creating a two-dimensional projection image based on an input from the information designating device 12.
And a defocus processing device 16 that performs defocus processing on a two-dimensional projection image (original image) based on inputs from the information specifying device 12 and the image creating device 14.

The defocus processing device 16 calculates the distance from the focal plane to the object on the basis of the input from the information designating device 12, a focus-object depth calculation unit 18, the depth calculation unit 18, and the image creation. The device 14 and a blurring processing unit 22 that performs blurring processing (filtering processing) based on an input from a processing number input device 20 separately provided, and the processed image created by the blurring processing unit 22 is an image storage device. It is designed to be stored in 24.

Hereinafter, each of the above processing devices will be sequentially described in detail.

The shape input device 10 is a device for inputting the shape data of an object, and the surface forming the object is represented by a collection of vertex information (X, Y, Z coordinates). Color information is also specified here.

The object position / light / camera / focal position information position designation device 12 includes the position of the object shape input from the shape input device 10, the position, direction and color of the light.
It has a function of designating light information such as intensity, camera information such as camera position, orientation, angle of view, and aspect ratio, and a focus position. The information can be specified while looking at the screen.

The two-dimensional projection image creating apparatus 14 places a viewpoint, a gazing point, and a light source at predetermined coordinate positions, calculates the brightness of a minute portion of each surface forming an object, and performs hidden surface processing,
It has a function of generating a shaded two-dimensional image (original image).

The defocus processing device 16 has
The two-dimensional image input from the three-dimensional projection image creating device 14, the object shape, the object position, the light information, the camera information input from the information designating device 12, and the number of processing times input from the processing number input device 20 are read. Then, the focus-object depth calculation unit 18 and the blurring processing unit 22 have a function of executing a defocus process described later.

The focus-object depth calculator 18 calculates the distance (depth difference) from the focus plane to the object for each pixel forming the object drawn in the original image. Figure 2
Shows a method of calculating the depth information, in which the distance dmax from the viewpoint to the focal point is 1.0.
In this case, the relative value of the depth (distance) d from the focus to the object P is calculated as a value df normalized by the following equation (1). Note that depth information is set to 1.0 for objects whose calculation results are 1.0 or more.

Df = d / dmax (1) where d: depth from focus to object dmax: depth between viewpoint and focus df: normalized focus-object depth information

In the above equation (1), the closer the object P is to the focus, the more df
The value of becomes closer to 0.0 and is in focus.
It means that the value of becomes close to 1.0 and blurs.

The depth information df is calculated for each pixel for use in the next filtering process.

In the blurring processing section 22, the depth calculation section 1
Based on the depth information df for each pixel calculated in 8,
Blur processing is performed by the following method. Here, a case will be described as an example where filtering processing is performed with a kernel size of 5 × 5.

The filter used is 5 as shown in FIG.
It consists of a kernel consisting of squares corresponding to × 5 pixels,
Each kernel is weighted by the weighting factor indicated by fw _ij in the figure. Here, 1 ≦ i ≦ 5 and 1 ≦ j ≦ 5.

The weighting coefficient fw _ij is calculated as follows. Center position of filter (i = 3, j = 3)
Corresponds to the pixel position to be processed.

Further, assuming that the depth information calculated for the target pixel by the equation (1) is df ₃₃ , the weighting coefficient fw ₃₃ of the kernel at the filter center is given by the following equation (2).

Fw ₃₃ = 1.0-df ₃₃ (2)

The weighting factor for the kernel other than the center, i.e., i = 3 and j = 3, is given by the following equation (3).
Set as fw _ij . However, in this equation (3), i = j
Not used when = 3.

Fw _ij = {(1.0−fw ₃₃ ) / (5 × 5-1)} × df _ij (3) (1 ≦ i ≦ 5, 1 ≦ j ≦ 5)

In the above equations (2) and (3), df ₃₃ = 0, that is, fw ₃₃ = 1 when the object P formed by the central pixel is at the focal position (when the object is in the best focus). .0
Then, all the weighting factors around it become 0, which means that no blur is applied.

As the weighting coefficient actually set in the filter, a value obtained by normalizing the weighting coefficient obtained by the above equations (2) and (3) by the following equation (4) is used. In the formula (4), for convenience,
Weighted coefficients after normalization are shown using the same symbols fw _ij as before normalization.

Fw _ij = fw _ij / ΣΣfw _nm (4)

When the filtering process is executed using the filter shown in FIG. 3, the obtained pixel value fp is expressed by the following equation (5).

Fp = p ₁₁ × fw ₁₁ ＋ p ₁₂ × fw ₁₂ ＋ p ₁₃ × fw ₁₃・ ・ ・ ・ P ₅₄ × fw ₅₄ ＋ p ₅₅ × fw ₅₅ (5) where fp is the result of the filtering process Pixel value of: p: pixel value of the two-dimensional projection image (original image)

When the filtering process is performed by using the above filter, the kernel at the center of the filter is set to the target pixel, and therefore, it cannot be directly applied to the image peripheral edge portion and the pixels inside thereof. Therefore, it is assumed that the same information as the peripheral edge pixels shown by 1 to 13 in the real image data (real area) shown by the solid line in FIG. 4 exists in the outer virtual area (shown by the chain double-dashed line). The filtering process is performed.

The processing number input device 20 is a device for inputting the number of times of performing the filtering process, and the filtering process of the number of times specified here is performed by the blurring processing unit 2.
It is executed in 2. When the filtering process is designated a plurality of times, the same filtering process is performed on the processed image, so that the blur can be strongly changed to an arbitrary degree. The number of times of filtering processing can be arbitrarily set as necessary.

Next, the operation of this embodiment will be described with reference to the flowchart of FIG.

First, when the shape input device 10 outputs information about the shape and color of an object to the position information designating device 12 (step 110), the position information designating device 12 outputs the information of the object based on the inputted information. A position, a light position, etc., a camera position, etc., and a focus position are designated, and the designation information is output to the two-dimensional projection image creation device 14 (step 112).

The image creating device 14 creates a two-dimensional image based on the input information from the position information designating device 12 (step 114). Next, in the defocus processing device 16, the focus-object depth calculation unit 18 calculates the difference in depth between the focus position and the object based on the depth information between the camera and the focus position that is input from the position information specifying device 12. (Step 116).

Next, the blurring processing number n is input to the blurring processing unit 22 from the processing number input device 20 (step 118). The blurring processing unit 22 creates the filter shown in FIG. 3 using the depth difference calculated by the focus-object depth calculating unit 18, and then the blurring processing unit 22 inputs the filter from the image creating apparatus 14. The blurring process (filtering process) is performed n times based on the information (step 120), and the processed image created as a result is stored in the image storage device 24.

According to the present embodiment, by performing the above-described series of processing, in three-dimensional CG image production, there is a focus position by blurring the image by filtering processing based on the distance information from the focus position of the object. That is, since it is possible to produce an image with defocus, it is possible to easily obtain a defocus effect.

Further, by performing the filtering process as a post-process on the two-dimensional original image created by rendering, the intensity of defocus can be easily changed, so that an arbitrary perspective feeling can be obtained in a short time. Can be set to an image.

The present invention has been specifically described above, but the present invention is not limited to the above-mentioned embodiments, and various modifications can be made without departing from the scope of the invention.

For example, the method of creating a two-dimensional original image is not limited to the ray tracing method, but pre-produced object shapes, color information, camera information, light information, mapping information, etc. are read and each of these is read. A rendering process such as radiocity, Z-buffer method, scan line method, etc., which is performed using information can be adopted.

Further, the filter used in the blurring process is not limited to the 5 × 5 kernel, and the number of kernels can be arbitrarily changed as required.

The weighting coefficient fw _ij used for the kernel is not limited to the one using the equations (2), (3) and (4), and can be changed as necessary.

[0047]

As described above, according to the present invention, 3
In three-dimensional computer graphics production, it is possible to easily set an arbitrary degree of blurring in a short time, so that it is possible to easily create a perspective image.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a defocus processing device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram for explaining a method of calculating depth information.

FIG. 3 is an explanatory diagram showing a configuration of a filter used in the above embodiment.

FIG. 4 is an explanatory diagram for explaining a filtering operation for an edge portion of an image.

FIG. 5 is a flowchart showing the flow of processing according to the above embodiment.

[Explanation of symbols]

 10 ... Shape input device 12 ... Position information designation device 14 ... Two-dimensional projection image creation device 16 ... Defocus processing device 18 ... Focus-object depth calculation unit 20 ... Processing number input device 22 ... Blur processing unit 24 ... Image storage unit

Claims (1)

[Claims] 1. A defocus processing device for blurring an image based on distance information from a focus position of an object in three-dimensional computer graphics production, wherein a two-dimensional original image produced by a rendering process, The function of calculating the depth difference between the focus position and the object based on the depth information between the camera and the focus position, and the original image based on the difference between the focus position of each pixel forming the object and the depth of the object. A defocus processing device, which further comprises a function of performing filtering processing.