JPH10214337A - Picture forming device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a picture forming device which quickly generates a picture to which a high semitransparent effect high is added. SOLUTION: A blend processing part 100 blends read color information CR of each picture element from a picture memory 120 and write color information CW based on a blend coefficient OP. A filter processing part 110 filters color information CB from the blend processing part 100 based on a filter coefficient F and writes obtained color information CF in the picture memory 120. The blend processing may be performed after the filter processing. Plural blend processing parts 100 and filter processing parts 110 may be provided, and at least one of the blend processing and the filter processing may be made ineffective. Thus, the number of times of access to the picture memory 120 is minimized even if one picture element is subjected to both the blend processing and the filter processing.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image generating apparatus capable of translucent display.

[0002]

2. Description of the Related Art Conventionally, various image generating apparatuses have been known for use in game machines, image generating tools, and the like. In such an image generation device, in order to realize a so-called virtual reality,
A major technical problem is how to improve the quality of an image. A technique called translucent processing is known as one of such image quality improving techniques.

In the translucent processing, a blending process for blending the color information of the translucent display object with the color information of the background written in the image memory and a filtering process for filtering the color information of the background written in the image memory are performed. By doing so, color information of each pixel is obtained. By performing the translucent processing, a translucent display object can be expressed, and the reality of an image can be significantly increased.

[0004] In order to obtain a higher translucent effect, it is desirable to perform both blend processing and filter processing on one display object. In this case, for example, the following method is employed. I do. That is, for example, first, a blending process is performed based on the color information read from the image memory, and the obtained color information is written to the image memory. Next, the written color information is read again from the image memory, a filtering process is performed based on the read color information, and the obtained color information is written again to the image memory.

However, in this method, it is necessary to access the image memory twice for reading and writing at the time of the blending process, and at the same time, to access the image memory twice for reading and writing at the time of the filtering process. Access is required. That is, access to the image memory is required four times in total. Since access to the image memory requires a lot of time, this method greatly hinders the processing speed. This is a serious problem particularly in an image generating apparatus that requires real-time processing.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned technical problems, and an object of the present invention is to provide an image generating apparatus capable of generating an image to which an advanced translucent effect is added at a high speed. Is to do.

[0007]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention relates to an image generating apparatus for performing a translucent process, comprising: reading color information for each pixel from image information storage means; Blend processing means for blending color information with a given blend coefficient; filtering the color information from the blend processing means with a given filter coefficient; and storing the obtained color information in the image information storage means. And a filter processing means for writing into Further, the present invention is an image generating apparatus that performs a translucent process, wherein a filter processing unit that filters read color information for each pixel from an image information storage unit based on a given filter coefficient; and And blend processing means for blending the color information and the given write color information based on a given blend coefficient, and writing the obtained color information to the image information storage means.

According to the present invention, it is possible to perform both the blending process and the filtering process on one pixel. As a result, it is possible to express a display object having both a blending property and a filtering property, and the reality of an image can be significantly increased. According to the present invention, the number of accesses to the image information storage unit can be minimized even when both the blending process and the filtering process are performed on one pixel. As a result, the processing can be speeded up, and an image generating apparatus that is particularly suitable for realizing the pipeline system translucent processing can be provided.

According to another aspect of the present invention, there is provided an image generating apparatus for performing a translucent process, comprising: a first processing unit for performing a first process based at least on color information read for each pixel from an image information storage unit; A second processing unit that performs a second process based on at least an output from the first processing unit, and a third processing unit that performs a third process based on at least an output from the second processing unit. ..., (N-1)
An N-th processing unit that performs an N-th processing based on at least an output from the (N ≧ 2) processing unit and writes the obtained color information into the image information storage unit; At least one of the processing means is a blend processing means for blending given two pieces of color information based on a given blend coefficient, and at least one of the first to Nth processing means is a given color information. An image generating apparatus, comprising: a filter processing unit that filters information based on a given filter coefficient.

According to the present invention, one pixel can be subjected to one or more blending processes and one or more filtering processes. Thereby, for example, even when a plurality of translucent display objects overlap on the background display object,
It is possible to generate an image full of realism while minimizing the number of accesses to the image information storage means.
Note that the first to N-th processes of the present invention include, besides blending and filtering, depth queuing,
Various processes such as a brightness process and a process using a reflectance parameter can be considered.

Further, the present invention is characterized in that it comprises means for invalidating at least one of the blending process performed by the blending unit and the filtering process performed by the filtering unit.

According to the present invention, the validity and invalidity of the blending process and the filtering process can be controlled. For example, when a display object to be processed has both a blending property and a filtering property, it can be dealt with by enabling both the blending process and the filtering process. If the display object to be processed has only one of the blending property and the filtering property, it can be dealt with by enabling one of the blending process and the filtering process and disabling the other.
If the display object to be processed is opaque and does not require translucent processing, it can be dealt with by disabling both the blend processing and the filter processing. For example, by enabling or disabling the translucent processing depending on the viewing direction of the display object, it is possible to express a display object such as a half mirror that becomes translucent or opaque depending on the viewing direction.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

First, blend processing (color interpolation processing) and filter processing will be described.

As shown in FIG. 1A, in the blending process, write color information (color information of a translucent display object) CW and color information (background color information) CR read from the image memory 220 such as a field buffer. , And the blend coefficient OP are input to the blend processing unit 200. Then, the blend processing section 200 writes the write color information CW and the read color information C
R is blended (interpolated) based on the blend coefficient OP, and the obtained color information CB is written to the image memory 220. Here, the blend processing unit 200 performs, for example, a blend process according to the following equation (1). RC _B = OP × R _CW + (1−OP) × R _CR G _CB = OP × G _CW + (1−OP) × G _CR (1) B _CB = OP × B _CW + (1−OP) × B _CR where R _CW, G _CW and B _CW are the RGB of the write color information CW
, R _CR , G _CR , and B _CR represent the RGB components of the read color information CR. R _CB , G _CB , B
_CB is the RG of the color information CB obtained by the blend processing unit 200
B represents each component. OP (0 ≦ OP ≦ 1), which is a blend coefficient, represents the opacity (opacity) of a translucent display object. The OP may be different for each of the RGB components. As the blend coefficient, various factors other than opacity, such as transparency (transparency), can be adopted.

On the other hand, as shown in FIG. 1B, in the filtering process, the color information C read from the image memory 220 is read.
R and the filter coefficient F are input to the filter processing unit 210. Then, the filter processing unit 210 reads the read color information C
R is filtered based on the filter coefficient F, and the obtained color information CF is written to the image memory 220. Here, the filter processing unit 210 performs, for example, a filter process according to the following equation (2). R _CF = F _R × R _CR G _CF = F _G × G _CR (2) B _CF = F _B × B _CR where R _CR , G _CR , and B _CR are RGB of the read color information CR.
Represents each of the components. R _CF , G _CF , and B _CF represent RGB components of the color information CF obtained by the filter processing unit 210. The filter coefficient F takes a different value for each of the RGB components. For the R component, F _R ,
For G component F _G, it takes the value of F _B is the B-component _{(0 ≦ F R, F G} , F B ≦ 1).

FIGS. 2A and 2B schematically show the concept of blending and filtering.

In the blending process, the color C1 of the translucent display object 20 itself and the color C2 of the background display object 22 are blended. More conceptually, in the blending process model, the translucent display object 20 can be considered as a transparent body including a plurality of color elements. These color elements emit light of the color C1. Then, the light of C2 color from the background display 22 passes through this transparent body at a given ratio. Therefore, the color C of light from the color element
1, the color C2 of the transmitted light is blended, and the blended color becomes a color that can be seen as a whole. The blend coefficient indicates the ratio of the area occupied by the color element to the area occupied by the transparent portion. If the area occupied by the color element is large, C
1 is greater, and the greater the area occupied by the transparent portion, the greater the effect of C2.

On the other hand, in the blending process, the background display 22
Is filtered by the translucent display object 20 to become C4. For example, if the color C3 of the background display 22 is white and the translucent display 20 is a filter that passes only the red component, the obtained C4 is red. When the translucent display object 20 is a filter that allows only a green component to pass, C4 is green.

For example, as shown in FIG. 2C, on a background display object 32 in which red characters 30 are drawn on a white base 28,
Consider a case in which a red translucent display object 34 is overlaid. Here, the color of the base 28 of the background display object 32 is set to (R, G, B) = (0.
5, 0.5, 0.5) (0 ≦ R, G, B ≦ 1), and the color of the character 30 and the translucent display 34 is (R, G, B) =
(0.5, 0, 0). Also, if the blend coefficient is OP =
0.5, and the filter coefficients are (F _R , F _G , F _B ) =
(0.5, 0, 0).

Then, in the case of the blending process, FIG.
As shown in (D), the color of the base 28 is (0.5, 0.2
5, 0.25), and the color of the character 30 is (0.5, 0,
0) (see equation (1) above). That is, the base 28 and the character 3
0 is a different color, and the presence of the character 30 can be recognized.

On the other hand, in the case of filter processing, FIG.
As shown in (E), the color of the base 28 is (0.25, 0,
0), and the color of the character 30 is also (0.25, 0, 0) (see the above equation (2)). That is, the base 28 and the character 30 have the same color, the presence of the character 30 cannot be recognized, and the character 30 appears to disappear.

Also, for example, when a translucent display object is superimposed on a black (0, 0, 0) background display object, in the filtering process, the color which is seen irrespective of the value of the filter coefficient is always black. . On the other hand, in the blending process, a color to which the color of the translucent display object is added becomes visible.

As described above, the obtained image effects are different between the blending process and the filtering process.
Existing translucent materials have both blend-like and filter-like properties. Therefore, in order to realize a more realistic translucent expression, it is desirable to perform both the blending process and the filtering process on one pixel. Then, for example, a method described below can be used to perform both the blend processing and the filter processing.

First, execution of blend processing is instructed by software. Then, in FIG. 1C, the software execution means 240 of the drawing processor 230
The color information CR is read from the image memory 220. Next, the execution means 240 calculates the CR, the write color information CW,
The output of the blend coefficient OP to the blend processing unit 200 and the execution of the blend processing are performed by the blend processing unit 20.
Indicate 0. Then, the blend processing unit 200 performs the blend processing represented by the above equation (1), and obtains the obtained color information CB.
Is returned to the drawing processor 230. Then, the drawing processor 230 writes the CB back to the image memory 220.

Next, when execution of the filter processing is instructed by software, the execution means 240
20 to read out the color information CR. This CR corresponds to the color information CB that has already been subjected to the blend processing by the blend processing unit 200. Next, the execution means 240
This CR and the filter coefficient F are combined with the filter processing unit 210
And instructs the filter processing unit 210 to perform the filter processing. Then, the filter processing unit 210
Performs the filter processing shown in the above equation (2), and returns the obtained color information CF to the drawing processor 230. Then, the drawing processor 230 writes the CF back to the image memory 220. In this way, it is possible to obtain color information on which both the blending process and the filtering process have been performed.

This method has the following advantages. In other words, according to this method, a free combination of processes is possible, such as performing a filtering process after the blending process, performing a blending process again, or inserting another process between the blending process and the filtering process. Becomes That is, versatility is high, and the degree of freedom of user operability can be increased.

However, this method has a disadvantage that the number of accesses to the image memory 220 is extremely large. That is, at the time of the first blending process, an access for reading the CR from the image memory 220 and an access for writing the CB obtained by the blending process to the image memory 220 are necessary. At the time of the next filter processing, access to read CR (= CB) after the blend processing from the image memory 220 and access to write the CF obtained by the filter processing to the image memory 220 are required. That is, a total of four accesses are required. Since the increase in the number of accesses to the image memory 220 greatly hinders the speeding up of the processing, this method is not suitable especially for an image generating apparatus that requires real-time image generation.

Therefore, the present embodiment employs a method as described below.

FIG. 3 shows an example of a functional block diagram of the image generating apparatus of this embodiment. The image generating apparatus includes a blend processing unit 100, a filter processing unit 110, and an image memory 12
Contains 0.

The blend processing unit 100 receives input of write color information (color information of a translucent display) CW, color information (background color information) CR read from the image memory 120 such as a field buffer, and a blend coefficient OP. You. The blend processing unit 100 is configured by hardware such as a linear interpolator, and blends the color information CW and CR based on the blend coefficient OP to obtain color information CB. That is, CB is obtained by a process as shown in the above equation (1). The obtained CB is output to filter processing section 110.

The filter processor 110 receives the color information CB and the filter coefficient F from the blend processor 100. The filter processing unit 110 is configured by hardware such as a multiplier, filters the color information CB from the filter processing unit 110 based on the filter coefficient F,
Find color information CF. That is, for example, the CF is obtained by the processing shown in the above equation (2), and the obtained CF is
Write to 0.

On the other hand, FIG. 4 shows another example of a functional block diagram of the image generating apparatus of the present embodiment. In FIG. 4, the arrangement of the blend processing unit 100 and the filter processing unit 110 is opposite to that in FIG. That is, the filter processing unit 110 performs a filtering process based on the read color information CR and the filter coefficient F, and outputs the obtained color information CF to the blend processing unit 100. The blend processing unit 100 includes a filter processing unit 11
A blending process is performed based on the color information CF from 0, the write color information CW, and the blend coefficient OP, and the obtained color information CB is written to the image memory 120.

In the case where both the blending process and the filtering process are performed, the number of accesses to the image memory 120 is four in the method described with reference to FIGS. On the other hand, according to the present embodiment, the number of accesses to the image memory 120 can be set to two. That is, for example, in FIG. 3, access for reading a CR from the image memory 120 first and access for writing the CF obtained by the filter processing unit 110 to the image memory 120 can be performed twice. Therefore, the processing can be remarkably speeded up as compared with the method described with reference to FIGS. 1A to 1C, and an image generating apparatus optimal for real-time image generation can be provided.

In particular, when performing translucent processing by a pipeline method, an increase in the number of accesses to the image memory 120 greatly hinders a high-speed processing. For example, in FIG. 3, it is assumed that the color information of the first, second,... In such a case, in the pipeline method, while the color information of the first pixel is being written into the image memory 120, the processing for the color information of the second pixel is performed by the filter processing unit 110 and the processing for the third pixel is performed. The processing is performed by the blend processing unit 100. As described above, in the pipeline system, one process is divided into a plurality of independent processing stages, arranged in a line, and each processing stage is executed in parallel, thereby substantially increasing the speed of the process. However, in the translucent processing, the color information read from the image memory is returned to the blend processing unit and the filter processing unit, so that the pipeline control is slightly complicated. In a method that requires two read accesses from the image memory when both the blending process and the filtering process are performed, the pipeline control is further complicated, and the speeding up of the process is further hindered. On the other hand, according to the present embodiment, even when both the blending process and the filtering process are performed, the number of times of read access from the image memory can be minimized, so that the control becomes much more complicated. The translucent processing can be pipelined without any need.

FIGS. 5 and 6 show functional block diagrams in the case where first to Nth processing units including a friend processing unit and a filter processing unit are provided.

[0037] In Figure 5, the first processing (blending process) 130 _-1 sought C ₁ based on the read color information CR from writing color information CW ₁ and the image memory 120 and the blending coefficient OP _1, the 2 processing (filter processing) unit 130
_-2 C ₁ and the filter coefficient F ₂ from the first processing unit 130 _-1
Request C ₂ based on and.

Further the third processing unit 130 _-3 obtains a C ₃ based on the writing color information CW ₃ and C ₂ and the blending coefficient OP ₃ from the second processing unit 130 _-2, the fourth processing unit 130 _-4
Determine the C ₄ on the basis of the C ₃ and the filter coefficient F ₄ from the third processing unit 130 _-3.

The (N-1) th (N ≧ 2) processing unit 13
0- _(N-1) is a value based on the write color information CW _N-1 , C _N-2 from the (N-2) th processing unit 130- _(N-2) and the blend coefficient OP _{N-1. N-1} is obtained, and the N-th processing unit 130- _N obtains C _N based on C _N-1 from the _(N-1) -th processing unit 130- _(N-1) and the filter coefficient F _N. writes the resultant C _N in the image memory 120.

In FIG. 6, the first processing unit _130-1 includes the read color information CR from the image memory 120 and the filter coefficient F
Seeking C ₁ based on ₁ and obtains a C ₂ based on the second processing unit 130 _-2 and writes color information CW ₂ and C ₁ and the blending coefficient OP ₂ from the first processing unit 130 _-1 .

Further the third processing unit 130 _-3 based on the C ₂ and the filter coefficient F ₃ from the second processing unit 130 _-2 C ₃
And the fourth processing unit _130-4 outputs the write color information CW ₄
When C ₃ and the blending coefficient OP ₄ from the third processing unit 130 _-3
Request C ₄ based on and.

The (N-1) th processing section 130- _(N-1) is the C _N-2 and the filter coefficient F _N-1 from the (N-2) th processing section 130- _(N-2). seeking C _N-1 based on the bets, C _N-1 and the blend from the processing section 130 _-N of the N-write color information CW _N (N-1) th processing unit 130 _{over (N-1)} C _N is obtained based on the coefficient OP _N, and the obtained C _N is written to the image memory 120.

As shown in FIGS. 5 and 6, by providing a plurality of stages of the blend processing unit and the filter processing unit, a plurality of blending processes and a plurality of filtering processes are performed on the pixel to be subjected to the translucent process. It becomes possible. For example, FIG.
As shown in the figure, two translucent display objects 42 and 44 are overlapped on the background display object 40, and the translucent display objects 42 and 44
When one blending process and one filtering process are performed for each of the above, after all, two blending processes and two filtering processes are required. Similarly, as shown in FIG. 7B, three translucent display objects 42, 4
When 4 and 46 overlap, three blending processes and three filtering processes are required. 3 and FIG. 4 in which the blend processing unit 100 and the filter processing unit 110 have only one stage, read access from the image memory 120 is performed twice in the case of FIG.
In the case of (B), three times are required, which hinders the speeding up of the processing. On the other hand, if a plurality of blend processing units and filter processing units are provided as shown in FIGS. 5 and 6, even if a plurality of blend processes and a plurality of filter processes are required, Can be minimized. For example, in the case of FIG. 7A, the blend processing unit and the filter processing unit have two stages (N = 2), and in the case of FIG. 7B, three stages (N = 3). , Image memory 12
The number of read accesses from 0 can be reduced to one.

When emphasizing downsizing of hardware at the expense of high-speed processing, it is desirable that the blend processing unit and the filter processing unit be provided one by one as shown in FIGS. . It is also possible to make the write color information CW _{1 to} CW _N , the blend coefficients OP _{1 to} OP _N , and the filter coefficients F _{1 to} F _{N the} same.

The first to Nth processing units 130 _{-1 to} 130 _{-1
In -N} , processes other than the blending process and the filtering process may be performed, and the order of performing the blending process and the filtering process may be various. For example, in FIG. 8, the blending process is performed in the second and sixth processing units 130 _-2 and 130 _-6 , and the filtering process is performed in the fourth processing unit _130-4 . On the other hand, in other processing units, various processes other than the blending process and the filtering process, such as a depth queuing process, a brightness process, a process based on a reflectance parameter, and the like, are performed.

In this embodiment, it is desirable to provide a means for invalidating at least one of the blending process performed by the blending process unit 100 and the filtering process performed by the filter processing unit 110 in FIGS. . Similarly, in FIGS. 5, 6, and 8, the first to Nth processing units 13
It is desirable to provide means for invalidating at least one of the blending process and the filtering process performed in 0 _{-1 to} 130 _-N . For example, in FIG. 7C, when the display object 52 overlapping the background display object 50 is a translucent display object having only a blending property, no filtering is necessary. Therefore, in this case, it is desirable to invalidate the filtering process. Conversely, if the display object 52 is a translucent display object having only a filter-like property, it is desirable that the blending process be unnecessary and the blending process be invalidated. If the display object 52 is an opaque display object, the translucent processing itself is not required.

FIGS. 9A and 9B show an example of a functional block diagram when at least one of the blending process and the filter process is invalidated. These correspond to the configuration of FIG.

In FIG. 9A, the blend processing unit 100
Enable terminals ENB, ENF
And enable terminals ENB, ENB
F is controlled by flags FLB and FLF. FLB = 1
In the case of, the processing of the blend processing unit 100 is enabled, and F
Invalid when LB = 0. When FLF = 1, the processing of the filter processing unit 110 is valid, and when FLF = 0, the processing is invalid.

In FIG. 9B, the pre-processing units 150, 1
By performing pre-processing on the blend coefficient OP and the filter coefficient F in 52, the validity and invalidity of the processing in the blend processing unit 100 and the filter processing unit 110 are controlled.

For example, when the flag FLB = 1, the preprocessing unit 150 outputs OP ′ = OP, and the blend processing unit 100 performs the blending process based on the OP ′. That is, the blending process becomes effective. On the other hand, when FLB = 0, the preprocessing unit 150 fixes OP ′ to 1. Fixing the blend coefficient OP 'to 1 is equivalent to disabling the blending process, as is apparent from the above equation (1).

When the flag FLF = 1, the pre-processing unit 152 outputs F ′ = F, and the filter processing unit 110 performs a filtering process based on F ′. That is, the filtering process becomes effective. On the other hand, when FLF = 0, the preprocessing unit 152 fixes the RGB components of F ′ to 1. Fixing each of the RGB components of the filter coefficient F 'to 1 is equivalent to disabling the filter processing, as is apparent from the above equation (2).

In FIGS. 4, 5, 6 and 8,
Although the blending process and the filtering process can be enabled and disabled in the same manner as described with reference to FIGS.
Here, the description is omitted.

The flags FLB and FLF can be set, for example, for each screen (for each frame), for each display object such as a polygon or a free-form surface, or for each part of the display object. For example, even on a screen in which color information, a blend coefficient, a filter coefficient, and the like are set to be the same, it is possible to make the appearance of the screen different by changing the settings of FLB and FLF.
Similarly, even for a display object or each part of the display object in which the color information, the blend coefficient, the filter coefficient, and the like are set to be the same,
By making the settings of FLB and FLF different, it is possible to make the appearance of the display object or each part of the display object different. Therefore, by using the flags FLB and FLF, the variety of obtained images can be remarkably increased, and a high-quality image can be generated with a small hardware scale and processing load.

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

For example, the blend processing and the filter processing performed by the blend processing means and the filter processing means are not limited to those described in the present embodiment. For example, the blend processing and the filter processing in consideration of the line-of-sight direction and light refraction are performed. And so on.

Also, the validity of the blending process and the filtering process,
The method of controlling the invalidation is not limited to the method shown in FIGS. 9A and 9B.

It is not necessary to write all of the color information for one frame in the image information storage means (image memory), and only a part of the color information for one frame may be written. For example, in an image generation tool or the like, an image to be generated may be divided into a plurality of parts, and an image of each of the divided parts may be generated. The present invention can be applied to such a case.

[0058]

[Brief description of the drawings]

FIGS. 1A, 1B, and 1C are diagrams for explaining an image generating apparatus capable of performing a blending process and a filtering process; FIG.

FIGS. 2A to 2E are diagrams for explaining the concept of blend processing and filter processing;

FIG. 3 is an example of a functional block diagram of the embodiment.

FIG. 4 is another example of a functional block diagram of the present embodiment.

FIG. 5 is a diagram illustrating first to fifth examples including a blend processing unit and a filter processing unit;
It is an example of a functional block diagram in the case of providing the Nth processing part.

FIG. 6 is a diagram illustrating a first example including a blend processing unit and a filter processing unit;
FIG. 14 is another example of a functional block diagram in the case where an Nth processing unit is provided.

FIGS. 7A to 7C are diagrams for explaining a plurality of times of blending processing, necessity of filter processing, and necessity of control of blending processing and validity / invalidity of filter processing;

FIG. 8 is a block diagram showing a first to a fifth embodiments including a blend processing unit and a filter processing unit;
FIG. 14 is another example of a functional block diagram in the case where an Nth processing unit is provided.

FIGS. 9A and 9B are examples of a functional block diagram in the case where the validity and invalidity of a blending process and a filtering process are controlled.

[Explanation of symbols]

REFERENCE SIGNS LIST 100 blend processing unit 110 filter processing unit 130 _{-1 to} 130 _-N first to Nth processing units 120 image memory (image information storage unit) 150, 152 preprocessing unit

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI H04N 9/74 G06F 15/72 350

Claims (4)

[Claims] 1. An image generating apparatus for performing translucent processing, comprising: blending read color information for each pixel from image information storage means and given write color information based on a given blend coefficient. Means for filtering the color information from the blend processing means based on a given filter coefficient, and writing the obtained color information to the image information storage means. . 2. An image generating apparatus for performing a translucent process, comprising: a filter processing unit configured to filter read color information for each pixel from an image information storage unit based on a given filter coefficient; And a blend processing unit for blending the color information and the given write color information based on a given blend coefficient and writing the obtained color information to the image information storage unit. . 3. An image generating apparatus for performing a translucent process, comprising: a first processing unit for performing a first process based at least on color information read out for each pixel from an image information storage unit; A second based at least on the output from the processing means;
, A third processing unit that performs a third process based on at least an output from the second processing unit,... (N-1) ( An N-th processing unit that performs an N-th processing based on at least an output from the processing unit of (N ≧ 2), and writes the obtained color information to the image information storage unit. At least one of the processing means is blend processing means for blending given two pieces of color information based on a given blend coefficient, and at least one of the first to Nth processing means is provided with given color information. An image generating apparatus characterized in that the image processing apparatus is a filter processing unit that performs filtering based on a given filter coefficient. 4. The apparatus according to claim 1, further comprising a unit that invalidates at least one of a blending process performed by the blending unit and a filtering process performed by the filtering unit. An image generation device.