JP4260747B2 - Moving picture composition method and scene composition method - Google Patents

Description

  The present invention relates generally to modeling languages for three-dimensional graphics, and more particularly to embedding images in scenes.

  In computer graphics, conventional real-time 3D scene rendering is based on computing a description of the scene's 3D geometry, which produces a representation of the image on a computer display. In the virtual reality modeling language (hereinafter abbreviated as VRML), many of the meanings commonly used in conventional three-dimensional applications are, for example, hierarchical transformation, light source, viewpoint, geometry, animation, fog It is a traditional modeling language that defines material properties, texture mapping, etc. The texture mapping process is generally used to apply externally supplied image data to a predetermined geometry in the scene. For example, image data supplied from the outside, video data supplied from the outside, or pixel data supplied from the outside can be applied to the surface by VRML. However, in VRML, a drawn scene cannot be used as a texture image that is declaratively mapped to another scene. In the declarative markup language, the semantics necessary to obtain the desired result are implicit, so the description of the obtained result is sufficient to obtain the desired result. Therefore, it is not necessary to prepare a processing procedure (that is, script description) for obtaining a desired result. As a result, it is desirable to be able to compose a scene using declarations. An example of a declarative language is a hypertext markup language (hereinafter abbreviated as HTML).

  Furthermore, it is desirable to generate a third surface by declaratively combining the two surfaces to which the image data is applied. It is also desirable to declaratively redraw image data applied to a surface to reflect the current state of the image.

  Conventionally, 3D scenes are drawn together, and the final frame rate for the viewer is determined by the worst performance determined by scene complexity or texture permutation. However, when different drawing speeds are applied to different elements on the same screen, the image quality is improved, and the viewing experience is more like watching a television, and is not a viewing experience like a web page.

  A computer system according to the present invention includes a computer and a computer program executed by the computer. The computer program includes a computer instruction for declaratively configuring a moving image using a first surface on which a first image is drawn and a second surface on which a second image is drawn, and a moving image Is generated by a declarative markup language, the first image is a three-dimensional representation of the object in space, the second surface is a protrusion having a three-dimensional representation in space, and the first image is the first The second image has a second opacity value, the moving image has a third opacity value, and the first image from the first surface and When a computer instruction that forms a moving image by combining the second image from the second surface and the first image and the second image are combined to obtain a traverse mat effect, A computer that moves the opacity value And an instruction. The first and second images are selected from the group consisting of video images, still images, animations, and scenes.

  Furthermore, a computer program in the computer system according to the present invention includes a computer instruction for drawing a first image on a first surface, a computer instruction for drawing a second image on a second surface, and a third And a computer instruction for declaratively rendering the image on the third surface. The first image is used as a texture for the third image, the second image is mixed with the texture to form a third image, the third surface is generated by a declarative markup language, The image is a protrusion having a three-dimensional representation in space, and the second image is a three-dimensional representation of the object.

  A moving image forming method for forming a moving image using a computer according to the present invention draws a first image on a first surface, draws a second image on a second surface, Generated by a declarative markup language, the first image is a three-dimensional representation of the object, the second surface is a protrusion having a three-dimensional representation in space, the first image and the second image, To declaratively compose a moving image, giving a first opacity value for the first image, giving a second opacity value for the second image, and a third opacity value for the moving image When the matte effect is obtained by combining the first image and the second image, the opacity value of the second image is moved.

  In a scene composition method for composing a scene using a computer according to the present invention, a first image is drawn on a first surface, a second image is drawn on a second surface, and the first image is drawn. Use as a texture for the scene, mix the texture with the second image, and draw the first scene on the third surface forming the first scene. The third surface is generated by a declarative markup language, the first image is a protrusion having a three-dimensional representation in space, and the second image is a three-dimensional representation of the object.

  The present invention provides a system and method for real-time composition and display of complex, dynamic, interactive experiences with an efficient declarative markup language. By using the surface composition, the creator can embed images or full motion video data in a place where a conventional texture map is used in his 3D scene. Further, the creator can use the drawing result of one scene description as an image that is texture-mapped to another scene. In particular, depending on the surface configuration, the result of any drawing application can be used as a texture in the creator's scene. Thereby, nested scenes can be drawn declaratively and scenes with component surfaces can be drawn at different drawing speeds.

  “Blendo” is an exemplary embodiment of the present invention, which controls animation and visible images, audio media, video media, animation, event data for the currently playing media asset. Allows temporal manipulation of media assets such as queuing. FIG. 1A is a block diagram showing the basic architecture of the blend. At the heart of the blend architecture is a core runtime module 10 (hereinafter referred to as the core), which is comprised of various elements of the application programmer interface (hereinafter abbreviated as API) and within the system 11. Provides an object model for the set of objects that exist in. During normal operation, the parser 14 parses the file into a raw scene graph 16 that is fed into the core 10 where the object is instantiated and a runtime scene graph is generated. The object can be a built-in object 18, an object 20 defined by the producer, a native object 24, or the like. These objects get a platform service 32 using a set of available managers 26. These platform services 32 include event processing, media asset loading, media playback, and the like. The object uses the rendering layer 28 to construct an intermediate or final image for display. The page integration component 30 is used to interface the blend to an external environment such as an HTML or XML page.

  The blend includes system objects for the set of managers 26. Each manager 26 provides a set of APIs for controlling some aspect of the system 11. Event manager 26D provides access to input device events resulting from user input or environmental events. The loading manager 26C facilitates blend file loading and native node execution. The media manager 26E provides the ability to load, control and play audio, image and video media assets. The rendering manager 26G makes it possible to generate and manage objects used for drawing a scene. The scene manager 26A controls the scene graph. The surface manager 26F allows the generation and management of the faces on which scene elements and other assets are combined. The thread manager 26B allows the producer to generate, control, and communicate between threads.

  FIG. 1B is a flowchart for conceptually explaining the flow of content by the blend engine. In step 50, the display begins with a source that includes a file or stream 34 (FIG. 1A) of content that is captured by parser 14 (FIG. 1A). The source format can be a text format such as native VRML, a native binary format, an XML-based format, or the like. In step 55, regardless of the source format, the source is converted to the raw scene graph 16 (FIG. 1A). The unprocessed scene graph 16 can represent initial values of fields as well as nodes, fields, and other objects in the content. It can also include object prototypes, external prototype references in stream 34, and descriptions of root statements.

  The top level of the unprocessed scene graph 16 includes nodes, top level fields and functions, prototypes, and routes included in the file. Blending allows the use of top-level fields and functions in addition to conventional elements. These are used to provide an interface to the external environment, such as an HTML page. They also provide an object interface when the stream 34 is used as external prototype content.

  Each unprocessed node contains a list of fields that are initialized within its context. Each raw field entry contains the name, type (if any), and data value of the field. Each data value includes a number, a string, an unprocessed node and / or an unprocessed field that can represent an explicitly entered field value.

  In step 60, a prototype is extracted from the top or raw scene graph 16 (FIG. 1A) and this prototype is used to generate a database of object prototypes accessible by the scene.

  The unprocessed scene graph 16 is then sent by a build traversal. During this movement, each object is constructed using the object prototype database (step 65).

  In step 70, a route within stream 34 is established. Thereafter, in step 75, each field in the scene is initialized. Field initialization is done by sending an initial event to the non-default field of the object. Since the scene graph structure is achieved by using node fields, in step 75, the scene hierarchy is also constructed. Events are emitted using in order traversal. The first node that appears lists the fields within the node. If the field is a node, that node is moved first.

  As a result, the nodes in that particular branch of the tree are initialized. One event is sent to the node field together with the initial value of the node field.

  After the fields for a particular node are initialized, the author adds initialization logic to the prototype object (step 80) to ensure that the node is fully initialized at the call time. Each of the above steps creates a root scene. In step 85, the scene is provided to the scene manager 26A (FIG. 1A) generated for that scene. In step 90, the scene manager 26A is used to perform rendering and action processing, either implicitly or under the control of the producer.

  The scene rendered by the scene manager 26A can be constructed using objects from the blend object hierarchy. Objects can take several functions from their parent object and extend or change their functions. There are objects at the bottom of the hierarchy. The two main object classes extracted from the object are nodes and fields. Nodes specifically include a rendering method, which is referred to as part of a render traversal. The data attribute of a node is called a field. Within the blend object hierarchy, there is a class of objects called timing objects, which will be described in detail below. Each code part below is exemplary. In addition, the line number of each code part simply represents the line number of a specific code part, and does not represent the line number in the original source code.

Surface objects
A surface object is a node of the surface node SurfaceNode type. The surface node SurfaceNode class is the base class for all objects that describe a two-dimensional image as an array of color, depth, and opacity (alpha) values. The surface node SurfaceNode is mainly used to provide an image used as a texture map. A movie surface MovieSurface, an image surface ImageSurface, a matte surface MatteSurface, a pixel surface PixelSurface, and a scene surface SceneSurface are derived from the class of the surface node SurfaceNode. The line number of each code part simply represents the line number for that code part, and does not represent the line number in the original source code.

MovieSurface
The following code part shows the node of the movie surface MovieSurface. The description of each field of the node of the movie surface MovieSurface is as follows.

1) MovieSurface: SurfaceNode TimedNode AudioSourceNode {
2) field MF String url []
3) field TimeBaseNode timeBase NULL
4) field Time duration 0
5) field Time loadTime 0
6) field String loadStatus “NONE”
}

  The movie surface MovieSurface node renders a movie on a surface by accessing an image sequence defining a movie. The parent class (parent class) of the timed node TimedNode of the movie surface MovieSurface defines which frames are drawn on the surface at a time. Movies can also be used as audio sources.

  In the second line of the code part, the URL field (of the multi-value field) provides a list of potential locations of the movie data relative to the face. This list is ordered so that element 0 describes the preferred source of data. If for some reason element 0 is not available or is in an unsupported format, the next element can be used.

  In the third line, when the time base timeBase field is designated, the time base timeBase field designates a node that provides timing information about the movie. In particular, the field of the time base timeBase gives the movie the information necessary to determine which frame of the movie is to be displayed on the surface at any given moment. If the timebase timeBase is not specified, the face displays the first frame of the movie.

  On the fourth line, when movie data is captured, a field of duration duration is set to the length of the movie by a node of the movie surface MovieSurface after a few seconds.

  In the 5th and 6th lines, the field of load time loadTime and the field of load status loadStatus provide information on availability of movie data from the node of the movie surface MovieSurface. The load status loadStatus can have five values, that is, “NONE”, “Request (REQUESTED)”, “Failure (FAILED)”, “Abort (ABORTED)”, and “Load (LOADED)”.

  “NONE” is the initial state. A “NONE” event is also sent when a node URL is cleared by setting the number of values to 0 or by setting the first string of URLs to an empty string. At this time, the surface pixels are set to black and opaque (that is, the color is 0, 0, 0, and the transparency is 0).

  A “REQUESTED” event is sent whenever a non-empty URL value is set. The face pixels do not change after the “REQUESTED” event.

  “FAILED” is sent after the “REQUESTED” event if the movie loading was not successful. This occurs, for example, when a URL refers to a file that does not exist or when the file does not contain valid data. The face pixels do not change after the “FAILED” event.

  The “ABORTED” event is sent when the current state is “REQUESTED” and the URL is changed. When the URL is changed to a non-empty value, a “request (REQUESTED)” event follows “ABORTED”. When the URL is changed to an empty value, a value of “NONE” comes after “ABORTED”. The surface pixels do not change after the “ABORTED” event.

  A “LOADED” event is sent when the movie is ready to be displayed. After that, an event of load time loadTime having a value matching the current time comes. The frame of the movie indicated by the timebase field is drawn on the screen. If the timebase timeBase is null, the first frame of the movie is drawn on the surface.

ImageSurface
The following code section shows the nodes of the image surface ImageSurface. The description of each field of the node of the image surface ImageSurface is as follows.

1) ImageSurface: SurfaceNode {
2) field MF String url []
3) field Time loadTime 0
4) field String loadStatus “NONE”
}

  The node of the image surface ImageSurface draws an image file on the surface. In the second line of the code part, the URL field provides a list of potential positions of the image data relative to the surface. This list is ordered so that element 0 describes the most preferred source of data. If for some reason element 0 is not available or is in an unsupported format, the next element can be used.

  In the third and fourth lines, a field for load time loadTime and a field for load status loadStatus provide information on availability of image data from a node of the image surface ImageSurface. The load status loadStatus can have five values, that is, “NONE”, “Request (REQUESTED)”, “Failure (FAILED)”, “Abort (ABORTED)”, and “Load (LOADED)”.

  “NONE” is the initial state. A “NONE” event is also sent when a node URL is cleared by setting the number of values to 0 or by setting the first string of URLs to an empty string. At this time, the surface pixels are set to black and opaque (that is, the color is 0, 0, 0, and the transparency is 0).

  A “REQUESTED” event is sent whenever a non-empty URL value is set. The face pixels do not change after the “REQUESTED” event.

  “FAILED” is sent after a “REQUESTED” event if image loading was not successful. This occurs, for example, when a URL refers to a file that does not exist or when the file does not contain valid data. The face pixels do not change after the “FAILED” event.

  The “ABORTED” event is sent when the current state is “REQUESTED” and the URL is changed. When the URL is changed to a non-empty value, a “request (REQUESTED)” event follows “ABORTED”. When the URL is changed to an empty value, a value of “NONE” comes after “ABORTED”. The surface pixels do not change after the “ABORTED” event.

  A “LOADED” event is sent when an image is drawn on a surface. After that, an event of load time loadTime having a value matching the current time comes.

Matte surface MatteSurface
The following code part shows the matte surface MatteSurface node. The description of each field of the matte surface MatteSurface node is as follows.
1) MatteSurface: SurfaceNode {
2) field SurfaceNode surface1 NULL
3) field SurfaceNode surface2 NULL
4) field String operation ““
5) field MF Float parameter 0
6) field Bool overwriteSurface2 FALSE
}

  The matte surface MatteSurface node uses the image composition operation to combine the image data from the surface surface1 and the surface surface2 into a third surface. The result of the compositing operation is calculated with the resolution of the surface surface2. When the size of the surface surface1 is different from the size of the surface surface2, the image data of the surface surface1 is zoomed up or down before the operation is performed, so that the size of the surface surface1 is equal to the size of the surface surface2.

  In the second and third lines of the code part, the surface surface1 and surface2 fields specify two surfaces that provide input image data for the compositing operation. In the fourth line, the operation field specifies a synthesis function to be executed on the two input surfaces. Possible operations are described below.

  The “Alpha Replace (REPLACE_ALPHA)” operation overwrites the alpha channel A of the surface surface2 with data from the surface surface1. If the surface surface1 has one component (greyscale intensity only), that component is used as the alpha (opacity) value. If the surface surface1 has two or four components (grayscale intensity + alpha value, ie RGBA), the alpha channel A is used to obtain the alpha value. If the surface surface1 has three components (RGB), the operation is undefined. This operation can be used to perform static or dynamic alpha masks on still or moving images. For example, the scene surface SceneSurface can draw a James Bond animated character on a transparent background. The alpha component of this image can be used as a mask shape for a video clip.

  The “alpha multiplication (MULTIPLY_ALPHA)” operation is the same as the alpha replacement operation except that the alpha value from the surface surface1 is multiplied by the alpha value from the surface surface2.

  The “cross fade (CROSS_FADE)” operation fades between two surfaces using a parameter value that controls the visible percentage of the two surfaces. This operation can dynamically fade between two still images or moving images. By moving the parameter value (5th line) from 0 to 1, the image of the surface surface1 fades to the image of the surface surface2.

  The “BLEND” operation combines image data from surfaces surface1 and surface2 using an alpha channel that controls the blending percentage from surface surface2. By this operation, the alpha channel of the surface surface2 can control the mixing ratio of the two images. By moving the alpha channel of the surface surface2 by drawing the scene surface SceneSurface or playing the movie surface MovieSurface, a complex and traverse matte effect is obtained. R1, G1, B1, and A1 represent the red, green, blue, and alpha values of the pixel on the surface surface1, and R2, G2, B2, and A2 represent the corresponding pixels red, green, blue, and alpha on the surface surface2. Are expressed as follows for the red, green, blue, and alpha components of the pixel.

Red = R1 * (1-A2) + R2 * A2 (1)
Green = G1 * (1-A2) + G2 * A2 (2)
Blue = B1 * (1-A2) + B2 * A2 (3)
Alpha = 1 (4)

  The “ADD” operation and the “SUBTRACT” operation add or subtract the color channels of the surface surface1 and the surface surface2. The resulting alpha value is equal to the alpha value of surface surface2.

  In line 5, the parameter field gives one or more floating point parameters that can change the effect of the compositing function. The specific interpretation of the parameter value depends on which operation is specified.

  In the sixth line, the field of the overwrite surface overwriteSurface2 assigns a new surface for the matte surface MatteSurface node to store the result of the compositing operation (overwriteSurface2 = FALSE), or the data stored in the surface surface2 Instructs whether to overwrite by overwrite (overwriteSurface2 = TRUE).

Pixel surface PixelSurface
The following code part shows a pixel surface PixelSurface node. The description of the field of the node of the pixel surface PixelSurface is as follows.
1) PixelSurface: SurfaceNode {
2) field Image image 0 0 0
}

  The node of the pixel surface PixelSurface draws an array of pixels specified by the user on the surface. In the second line, the image field describes pixel data to be drawn on the surface.

SceneSurface
The following code section shows a scene surface SceneSurface node. The description of each field of the node of the scene surface SceneSurface is as follows.
1) SceneSurface: SurfaceNode {
2) field MF ChildNode children []
3) field UInt32 width 1
4) field UInt32 height 1
}

  The node of the scene surface SceneSurface draws a specific child node on a surface of a specific size. The scene surface SceneSurface automatically redraws to reflect the current state of the child nodes.

  In the second line of the code part, the child field describes a child node to be drawn. Conceptually, the child field describes the entire scene graph that is drawn independently of the scene graph that contains the nodes of the scene surface SceneSurface.

  In the third and fourth lines, the width field and the height field specify the size of the surface in pixels. For example, if the width is 256 and the height is 512, the surface has an array of 256 × 512 pixel values.

  The node of the movie surface MovieSurface, the node of the image surface ImageSurface, the node of the matte surface MatteSurface, the node of the pixel surface PixelSurface, and the node of the scene surface SceneSurface are used for drawing the scene.

  At the top of the scene description, the output of the “top level surface” is mapped to the display. Rather than rendering the result on the display, the output can be generated on the surface using one of the surface nodes described above, with the scene rendered in three dimensions. Here, the output can be incorporated into a high quality scene structure as desired by the producer. Surface content is generated by drawing a scene description embedded in the surface, which may include color information, transparency (alpha channel), and depth as part of the image configuration of the surface. An image in this case is defined as including a video image, a still image, an animation, or a scene.

  Surfaces are also defined as internally supporting specialized conditions of various texture mapping systems rather than a general image management interface. As a result, any surface producer in the texture mapping system can be used as a texture by a three-dimensional drawing process. Examples of such surface producers include an image surface ImageSurface, a movie surface MovieSurface, a matte surface MatteSurface, a scene surface SceneSurface, and an application surface ApplicationSurface.

  The application surface ApplicationSurface maintains image data embedded and drawn by application processing, for example, image data such as a spreadsheet or a word processor drawn by a method similar to an application window in a conventional window system.

  By integrating surface models with rendering production and using textures, declarative authoring at different rendering speeds is possible. Conventionally, 3D scenes are drawn together, and the final frame rate for the viewer is determined by the worst performance resulting from scene complexity or texture replacement. In a continuous configuration of a framework in real time, a mechanism that makes drawing speeds different for different elements on the same screen can be obtained by abstracting surfaces. For example, it is possible to draw a web browser that draws at a low rate of 1 frame per second, but this is generated by other applications and the video frame rate displayed with the browser output is the full frame rate per second. This is possible only if it is maintained at 30 frames. When a web browser application draws a surface on itself, the screen compositor can quickly update the fully rendered image at the end of the web browser surface without being interrupted by the full motion video frame rate. Can be used as part to draw.

  FIG. 2A is a diagram for explaining a method of drawing a complicated portion 202 of the screen display 200 at a full motion video frame rate. FIG. 2B is a flowchart showing various operations when the screen display 200 including the complicated portion 202 is drawn at the full motion video rate. The screen display 200 is preferably displayed at 30 frames per second, but the portion 202 of the screen display 200 is too complex to display at 30 frames per second. In this case, in step 210 (FIG. 2B), the portion 202 is drawn on the first surface and stored in the buffer 204. In step 215, the screen display 200 including the portion 202 is displayed at 30 frames per second by using the first surface stored in the buffer 204. While the screen display 200 including the portion 202 is being displayed, the next frame of the portion 202 is drawn on the second surface and stored in the buffer 206 in step 220. Once the next frame of this portion 202 is available, the next update of the screen display 200 is performed using the second surface (step 225) and a further updated version of the portion 202 is available in the buffer 204. Continue this until. While the screen display 200 is displayed using the second surface, the next frame of the portion 202 is drawn on the first surface in step 230. When the drawing of the next frame of the portion 202 on the first surface is completed, the updated first surface is used to display the screen display 200 including the complex portion 202 at 30 frames per second.

  Nested scenes can be declaratively drawn by integrating a surface model with drawing generation and using textures. By recombining the sub-scenes drawn as images, open-end authoring becomes possible. In particular, by using sub-scenes that are animated and image mixed into a larger video context, a more appropriate image quality can be obtained as computer graphics for entertainment. For example, this method of blending images gives visual artists an alternative to the previous generation of rough, well-defined clipping by the window system.

  FIG. 3A shows a nested scene that includes animated sub-scenes. FIG. 3B is a flowchart illustrating operations performed when rendering the nested scene of FIG. 3A. In step 310, the background image displayed on the screen display 200 is drawn, and in step 315, the cube 302 is placed in the background image displayed on the screen display 200. The area outside the cube 302 is a part of the surface that forms the background of the cube 302 on the screen display 200. Surface 304 of cube 302 is defined as the third surface. In step 320, a movie is drawn on the third surface using the movie surface MovieSurface node. Thereby, the surface 304 of the cube 302 displays the movie drawn on the third surface. Surface 306 of cube 302 is defined as the fourth surface. In step 325, an image is drawn on the fourth surface using the node of the image surface ImageSurface. Thereby, the surface 306 of the cube 302 displays the image drawn on the fourth surface. In step 330, the entire cube 302 is defined as the fifth plane, and in step 335, the fifth plane is moved and / or rotated to play a movie on the plane 304 and a still image on the plane 305. The moving cube to be displayed is configured. By performing the processing procedure described above, it is possible to display different drawings on each surface of the cube 302. Steps 310 to 335 may be performed in any order, and all of steps 310 to 335 may be started simultaneously.

  Note that the present invention is independent of “Blendo” and can be part of another embodiment different from “Blend”. In the description of the present invention, the drawing of a three-dimensional scene is described. However, the present invention can be similarly applied to the drawing of a two-dimensional scene. The surface model allows authors to freely combine images and video effects with 2D and 3D geometric mapping and animation.

  Although the invention has been illustrated and described with respect to specific embodiments, it will be apparent to those skilled in the art that various changes and modifications can be made without departing from the features of the invention. Therefore, the scope of the claims includes all such changes and modifications as such changes and modifications are within the spirit and scope of the present invention.

It is a figure which shows the basic architecture of a blend (Blendo). It is a flowchart which shows the flow of the content by a blend engine. It is a figure which shows how two surfaces in one scene are drawn at different drawing speeds. It is a flowchart which shows the operation | movement performed when drawing two surfaces shown to FIG. 2A at different drawing speeds. It is a figure which shows the nested scene. 3B is a flowchart illustrating operations performed to render the nested diagram of FIG. 3A.

Claims (27)

In a computer system comprising a computer and a computer program executed by the computer,
The computer program is
A first surface the first image is drawn at a first rate, videos second image using a second surface drawn at the second speed, declaratively constituting a moving picture Computer instructions for image construction ;
The moving image is generated by a declarative markup language, the first image is a three-dimensional representation of the object in space, and the second surface is one surface of the three-dimensional representation of the object in space. The first image has a first opacity value, the second image has a second opacity value, and the moving image has a third opacity value; A computer instruction for constructing a moving image by combining the first image from the first surface and the second image from the second surface on a display;
A computer instruction for moving image configuration that moves the opacity value of the second image when the first image and the second image are combined to obtain a traveling mat effect,
The computer system according to claim 1, wherein the first and second images are selected from a group consisting of a video image, a still image, an animation, and a scene.   The computer system according to claim 1, wherein the first image is drawn with a first two-dimensional pixel array, and the second image is drawn with a second two-dimensional pixel array.   The computer system according to claim 1, wherein the moving image is configured by a declarative instruction.   The computer system according to claim 3, wherein the moving image is configured in real time.   The computer system according to claim 2, wherein the moving image is configured by a declarative instruction.   The computer program overwrites the second opacity value with the first opacity value when the moving image is formed by combining the first image and the second image. The computer system according to claim 1, further comprising instructions.   2. The computer program according to claim 1, further comprising a computer instruction for multiplying the first opacity value by the second opacity value to obtain the third opacity value. Computer system. In a computer system comprising a computer and a computer program executed by the computer,
The computer program is
Computer instructions for scene composition for rendering a first image on a first surface at a first speed ;
Computer instructions for scene composition for rendering a second image on a second surface at a second speed ;
A computer instruction for scene composition for declaratively rendering a third image on the third surface;
The first image is used as a texture for the third image, the second image is mixed with the texture on the display to form the third image, and the third surface is declarative generated by a markup language, the first surface is one surface of the three-dimensional representation of an object in space, the computer system characterized in that said second image is a three-dimensional representation of the object .   9. The computer system according to claim 8, wherein the first, second and third images are drawn by a declarative instruction from a user.   The computer system according to claim 8, wherein the second image changes with time.   The computer program has a computer instruction for declaratively drawing a fourth image on a fourth surface, and the third image is mixed in the fourth image to be included in the fourth image. 9. The computer system according to claim 8, wherein a sub-scene is formed on the computer.   12. The computer system of claim 11, wherein the second image in the sub-scene changes to reflect a change in the second image of the second surface.   9. The computer system according to claim 8, wherein the first and second images are selected from the group consisting of a video image, a still image, an animation, and a scene.   9. The computer system according to claim 8, wherein the drawing of the first, second, and third images is performed recursively. In a moving image composition method for forming a moving image using a computer,
Drawing a first image on a first surface at a first speed ;
Drawing a second image on a second surface at a second speed ;
The moving image is generated by a declarative markup language, the first image is a three-dimensional representation of the object, the second surface is one surface of the object having a three-dimensional representation in space, Combining the first image and the second image on a display to construct the moving image declaratively;
Giving a first opacity value of the first image,
Giving a second opacity value for the second image,
Give a third opacity value for the video,
A moving image constructing method for moving an opacity value of the second image when a traveling mat effect is obtained by combining the first image and the second image.   The method according to claim 15, wherein the first and second images are selected from a group consisting of a video image, a still image, an animation, and a scene.   The method according to claim 16, wherein the scene includes at least one of a group consisting of a video image, a still image, an animation, and a scene. Furthermore, the first image is drawn with a first two-dimensional pixel array,
The moving image composition method according to claim 15, wherein the second image is drawn by a second two-dimensional pixel array.   16. The moving picture composition method according to claim 15, further comprising providing a declarative instruction for constructing the moving picture.   The method according to claim 15, wherein the moving image is configured in real time.   16. The moving image composition method according to claim 15, wherein the second opacity value is overwritten with the first opacity value.   The method of claim 15, further comprising multiplying the first opacity value by the second opacity value to obtain the third opacity value. In a scene composition method for composing a scene using a computer,
Drawing a first image on a first surface at a first speed ;
Drawing a second image on a second surface at a second speed ;
The first image is used as a texture for the first scene, the texture is mixed with the second image on the display , and the first scene is formed to form the first scene. Draw on the surface,
Said third surface is generated by the declarative markup language, the first surface is one surface of the three-dimensional representation of an object in space, the second image is a three-dimensional representation of the object A scene composition method characterized by being.   24. The scene composition method according to claim 23, further comprising declarative instructions for drawing the first and second images and the first scene.   The scene composition method according to claim 23, wherein the second image changes with time.   The scene composition method according to claim 23, wherein the first and second images are selected from a group consisting of a video image, a still image, an animation, and a scene.   The scene composition method according to claim 23, wherein the drawing of the first and second images and the third scene is performed recursively.

Images