JP2005250748A - Video composition device, video composition program, and video composition system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a video synthesizing apparatus, a video synthesizing program, and a video synthesizing system that can output a synthesized video in which a person behaves naturally with respect to a CG image without wearing a sensor and special glasses. .
A video synthesizing apparatus 1 includes a video portion in which a three-dimensional model modeled by a three-dimensional model modeling apparatus 2 according to three-dimensional CG data and a person photographed by a photographing camera 4 are projected. The CG image drawn according to the 3D CG data is superimposed and output. The 3D CG data storage means 3, the 3D CG data output control means 5, the rendering means 7, and the 3D model image A partial detection means 9 and a replacement means 11 were provided.
[Selection] Figure 1

Description

  The present invention relates to a video synthesis device, a video synthesis program, and a video synthesis system that synthesize a three-dimensional CG with a shot video.

  Conventionally, as a technique for arranging a 3D CG, which is a 3D image drawn by a computer, in a live-action image taken by a shooting camera, there are a virtual studio technique and a mixed reality technique (see, for example, Patent Documents 1 and 2). ).

  In the virtual studio technology, a studio set is drawn as a CG image by a computer (virtually assembled), and a live-action image of a person (performer, for example, an announcer) is drawn in the drawn CG image (virtual model). By inserting (arranging and synthesizing), it is possible to create an image as if a person (performer) is in a real studio set even though the studio set does not actually exist. It is a technology that can.

  The mixed reality technology is a technology that seamlessly fuses a virtual world such as a CG image and a real world such as an actually captured image in real time. As one aspect of this mixed reality technology, a person (user) A sensor capable of acquiring a position and a line-of-sight direction, special glasses capable of presenting a synthesized video obtained by synthesizing a CG image to a live-action video to the person (user), and an arrangement indicating a position where the CG image is arranged in an actual space Some use a computer that stores data.

  In one aspect of this mixed reality technology, when a person (user) wears a sensor and special glasses and performs an arbitrary operation, the sensor outputs sensor data that changes according to the arbitrary operation to a computer. Subsequently, the computer generates a CG image that can be seen from the position and eye direction of the person (user) from time to time based on the input sensor data and arrangement data, and the generated CG image is displayed by using special glasses. It is presented to a person (user) by arranging it on a live-action video in real time.

As a result, the person (user) can feel as if a CG image is present at a specific location in the actual space.
Japanese Patent Laid-Open No. 11-261888 (paragraphs 0009 to 0012, FIG. 1) Japanese Patent Laying-Open No. 2003-303356 (paragraphs 0012 to 0022, FIG. 1)

  However, in the conventional virtual studio technology, even if a studio set is generated as a CG image (virtual model), a person (performer) uses a television monitor to display a composite image in which the CG image is embedded in a live-action image. For confirmation, it is impossible to accurately grasp the presence or arrangement position of a CG image, and natural behavior cannot be performed on the CG image. For example, a line of sight when paying attention to a position where a CG image exists In other words, if a plurality of persons (performers) are paying attention to one specific place in the CG image (sharing of eyes) or using a pointing stick or the like, the insertion position is unnatural. There is a problem of becoming.

  In addition, in the conventional mixed reality technology, natural behavior can be performed on a CG image arranged in an actual space, but a person (user) must wear a sensor and special glasses. When this conventional mixed reality technology is used for program production or the like, there is a problem that it is difficult to produce a program because a person who cannot confirm the face wearing these sensors and special glasses will appear. .

  Furthermore, when a model explained by a person (performer) is actually produced and used in a program, the model must be produced with the actual scale down to an appropriate size. When an aircraft accident is handled in a program, an elaborate model of the aircraft may be required. In such a case, there is a problem that it takes time to create an elaborate model.

  Therefore, the present invention solves the above-described problems, eliminates the need for wearing sensors and special glasses, can deal with urgent news programs without creating elaborate models, An object of the present invention is to provide a video composition device, a video composition program, and a video composition system that can output a composite image in which a person (performer) behaves naturally.

  In order to solve the above-described problem, the video composition device according to claim 1 is configured such that the three-dimensional model is projected in the image obtained by photographing the three-dimensional model and the person modeled by the three-dimensional model modeling unit according to the three-dimensional CG data. A video synthesizing apparatus for superimposing and outputting a 3D CG drawn in accordance with the 3D CG data on a video portion, and including a 3D CG data storage unit, a 3D CG data output unit, a rendering unit, The three-dimensional model image part detecting means and the replacing means are provided.

  According to such a configuration, the video composition device causes the 3D CG data output means to output the 3D CG data stored in the 3D CG data storage means to the 3D model modeling means. The three-dimensional CG data includes a coordinate group indicating a shape for drawing three-dimensional computer graphics, texture information indicating the texture of the three-dimensional CG to be drawn, glow information indicating the lightness and darkness and color of the drawn CG image, and the like. Is included.

  In addition, the video composition device uses the rendering unit to position the shooting unit (for example, coordinates xyz), the direction in which the shooting unit is shooting, and the zoom amount when the shooting unit is shooting the three-dimensional model ( The three-dimensional model is converted into a three-dimensional CG based on the shooting information including the zoom ratio), the arrangement information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data stored in the three-dimensional CG data storage means. In this case, a CG image (two-dimensional CG image) that will be photographed by the photographing means when being arranged is drawn (generated) every moment. That is, a two-dimensional CG image viewed from the direction of the photographing means of the model modeled by the three-dimensional model shaping means is sequentially generated as the photographing information including the position, direction, and zoom amount of the photographing means that is photographing is changed. Will be.

  Then, the video composition device detects the video part in which the three-dimensional model in the live-action video shot by the photographing unit is projected by the three-dimensional model video part detecting unit, and the detected video part is detected by the replacement unit. The CG image drawn by the rendering means is replaced and output. That is, a CG image (a two-dimensional CG image) whose shape (appearance) changes with changes in shooting information including the position, direction, and zoom amount of the shooting unit drawn by the rendering unit is converted into a three-dimensional model by the replacement unit. It will be replaced with the video part detected by the video part detection means.

  According to a second aspect of the present invention, in the video synthesizing apparatus according to the first aspect, the three-dimensional model forming means uses a principle of a thin film stacking method in a rapid prototyping technique.

  According to such a configuration, the video synthesizing apparatus uses the principle of the thin film stacking method in the rapid prototyping technique for the output 3D model forming means for outputting the 3D CG data. The three-dimensional model representing the shape is formed by cardboard or the like.

  The video composition program according to claim 3, in a video portion in which the three-dimensional model is projected in a video obtained by photographing the three-dimensional model and the person modeled by the three-dimensional model modeling means according to the three-dimensional CG data, by the photographing means, In order to superimpose and output three-dimensional CG drawn according to the three-dimensional CG data, the computer is configured to function as a three-dimensional CG data output unit, a rendering unit, a three-dimensional model video portion detection unit, and a replacement unit.

  According to this configuration, the video composition program causes the three-dimensional CG data output unit to output the three-dimensional CG data stored in the three-dimensional CG data storage unit to the three-dimensional model modeling unit. Also, the video composition program is stored in the 3D CG data storage unit by the rendering unit, the shooting information including the position, direction and zoom amount of the shooting unit, the location information indicating the position where the 3D model is placed, and the 3D CG data storage unit. Based on the existing 3D CG data, the CG image (2D CG image) that will be captured by the imaging means when the 3D model is replaced with the 3D CG is arranged (every time) drawn (generated) ) The video composition program detects the video portion in which the three-dimensional model in the live-action video shot by the photographing means is projected by the three-dimensional model video portion detecting means, and the detected video portion is detected by the replacing means. The CG image (two-dimensional CG image) drawn by the rendering means is replaced and output.

  The video composition system according to claim 4, wherein the three-dimensional model is displayed in a video portion where the three-dimensional model is projected in a video obtained by photographing a three-dimensional model and a person who explains using the three-dimensional model. A video composition system that replaces and outputs a CG image drawn according to CG data, and includes a three-dimensional CG data storage unit, a three-dimensional CG data output unit, a three-dimensional model modeling unit, a photographing information acquisition unit, a rendering unit, The three-dimensional model image part detecting means and the replacing means are provided.

  According to such a configuration, the video composition system causes the three-dimensional CG data output means to output the three-dimensional CG data stored in the three-dimensional CG data storage means to the three-dimensional model shaping means. In addition, the video composition system forms a three-dimensional model by the three-dimensional model forming means, and the photographing information acquisition means sets the position and direction (for example, the coordinate xyz and direction) where the photographing means is installed and the photographing means converts the three-dimensional model. Shooting information including a zoom amount (zoom rate) at the time of shooting is acquired. Then, the video composition system uses the rendering unit to acquire the shooting information acquired by the shooting information acquisition unit, the arrangement information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data storage unit. Based on the CG data, a CG image (two-dimensional CG image) that will be photographed by the photographing means when the three-dimensional model is replaced with the three-dimensional CG is arranged (generated) every moment. Then, the video composition system detects the video portion in which the three-dimensional model in the live-action video photographed by the photographing means is projected by the three-dimensional model video portion detection means, and detects the video portion detected by the replacement means. The CG image drawn by the rendering means is replaced and output.

  According to the first, third, and fourth aspects of the invention, the person does not need to wear a sensor and special glasses, and can operate while looking at the actually modeled three-dimensional model. In addition, since the three-dimensional model is replaced with the three-dimensional CG, it is possible to output a composite image obtained by performing a natural behavior with respect to the three-dimensional CG (CG image).

  According to the second aspect of the present invention, if the three-dimensional CG data is prepared in advance, a three-dimensional model can be easily obtained. For example, a real-time live broadcast program can be quickly handled.

Next, embodiments of the present invention will be described in detail with reference to the drawings as appropriate.
(Configuration of video composition device)
FIG. 1 is a block diagram of a video composition system. As shown in FIG. 1, a video composition system A includes a video composition device 1, a three-dimensional model shaping device 2 (three-dimensional model shaping means), a photographing camera 4 (photographing means), and a sensor 6 (photographing information acquisition means). It consists of and. Prior to the description of the video composition device 1, another configuration will be described.

  Based on the input data (three-dimensional CG data), the three-dimensional model forming apparatus 2 uses the principle of the thin film stacking method (LOM method) of rapid timing technology to make a sheet-like material such as thin paper (corrugated cardboard paper or A three-dimensional model is formed by drawing a contour line on a material (such as a piece of paper) and sequentially stacking the cut material while cutting the contour line of the material with a laser or a knife. In addition, the raw material (material) used with this three-dimensional model modeling apparatus 2 is made into a single color so that it can detect with the three-dimensional model image | video part detection means (after-mentioned) of the image | video composition apparatus 1. FIG.

  In the thin film lamination method (LOM method) of the rapid timing technology, a thin film-like material is cut out based on the data related to the XY plane in the input data, and this cut out material is the height of the input data. Lamination is performed based on the data regarding the direction to create a “lumb” (master model) of a material having a specific shape, and is applied to the production of a mold or the like.

  Specifically, the three-dimensional model forming apparatus 2 may use a commercially available RP system (rapid prototyping system) applied to sand mold casting and precision casting, and a thin film used in the RP system. Instead of handling only this piece of paper, it may be improved so that a material such as corrugated cardboard can be handled in order to form a three-dimensional model at high speed.

  However, when a commercially available RP system is used as it is as the three-dimensional model forming apparatus 2, this RP system generally takes time to form a three-dimensional model and the like, and it is necessary to create a three-dimensional model used in emergency programs and the like. Is assumed to be incompatible. Therefore, it is assumed that a commercially available RP system is applied to programs other than emergency programs.

  The three-dimensional model formed by using the material such as corrugated cardboard for modeling at high speed by the three-dimensional model modeling apparatus 2 cuts off unnecessary portions of the material such as corrugated cardboard according to the input data (three-dimensional CG data). (Scraped out), and the material such as corrugated paper from which unnecessary portions are cut off is stacked several times.

  That is, in this video composition system A, since the principle of the thin film stacking method in the rapid prototyping technology is used for the output 3D CG data output device 2 for outputting the 3D CG data, the 3D CG data is prepared in advance. If this is done, a three-dimensional model approximating the shape of a three-dimensional CG will be modeled at high speed using corrugated cardboard or the like, and for example, it can quickly respond to a real-time live broadcast program.

  The photographic camera 4 shoots a three-dimensional model formed by the three-dimensional model forming apparatus 2 and a person who performs an action (acting and commentary) while looking at the three-dimensional model, and has photographed the three-dimensional model and the person. The live-shooting signal is output to the video composition device 1. The photographing camera 4 is employed by a general broadcasting station or the like.

  The sensor 6 includes a position (for example, coordinates xyz) where the shooting camera 4 is installed, a direction in which the shooting camera 4 is shooting, and a zoom amount (zoom rate) when the shooting camera 4 is shooting a three-dimensional model. Detecting (acquiring) photographing information including In this embodiment, the sensor 6 includes the following elements that detect the position, direction, and zoom amount (zoom rate).

  A sensor 6 that detects a position where the photographing camera 4 is installed is a transmitter (not shown) that is installed in the photographing camera 4 and transmits a signal including at least one of magnetism, infrared rays, and ultrasonic waves. A position detector (not shown) that detects a position based on a signal transmitted from the transmitter is configured.

  The sensor 6 that detects the direction in which the photographing camera 4 is photographing is a gyro compass or the like that detects the direction.

  The sensor 6 for detecting the zoom amount (zoom rate) of the photographing camera 4 is attached to an external interface (not shown) of the photographing camera 4 and is output from the external interface (not shown). This is a detection device for detecting the rate.

Next, the configuration of the video composition device 1 will be described.
The video synthesizing device 1 synthesizes the drawn three-dimensional CG with the real video signal photographed by the photographing camera 4, and outputs the synthesized three-dimensional CG as a synthesized video signal. The three-dimensional CG data storage means 3, A three-dimensional CG data output control means 5 (three-dimensional CG data output means), a rendering means 7, a three-dimensional model video portion detection means 9, and a replacement means 11 are provided.

  The three-dimensional CG data storage means 3 stores three-dimensional CG data that is data for drawing a three-dimensional CG, and is configured by a general recording medium such as a hard disk. This three-dimensional CG data storage means 3 sends the stored three-dimensional CG data to the three-dimensional model modeling apparatus 2 outside the video composition apparatus 1 according to the control signal output from the three-dimensional CG data output means 5. In addition to outputting, it also outputs to the rendering means 7.

  The three-dimensional CG data includes a coordinate group indicating a shape for drawing three-dimensional computer graphics, texture information indicating the texture of the three-dimensional CG to be drawn, and glow information indicating the lightness and darkness and color of the three-dimensional CG to be drawn. Etc. For example, when the 3D CG to be drawn is a plurality of buildings such as 1/100 scale high-rise buildings, the coordinate group has 8 shapes (vertex of the quadrangular columns) indicating at least the shape of the building, for example, a quadrangular column. The texture information indicates the surface material of the building, for example, glass and concrete, and the glow information indicates the color of the building, for example, light blue.

  The three-dimensional CG data output control means 5 operates the input means (not shown) by the operator of the video synthesizer 1 and, based on the input operation signal, the three-dimensional CG data storage means 3 to form a three-dimensional model. A control signal for causing the apparatus 2 and the rendering means 7 to output 3D CG data is output to the 3D CG data storage means 3.

  The rendering means 7 includes shooting information including the position, direction, and zoom amount (zoom rate) of the shooting camera 4 detected by the sensor 6, and arrangement information indicating the position where the three-dimensional model is arranged, Based on the three-dimensional CG data stored in the three-dimensional CG data storage unit 3, a CG image (two-dimensional CG image) viewed from the photographing camera 4 is drawn (generated) each time. The drawing of the CG image by the rendering means 7 is generated by a general CG drawing method. This arrangement information is stored in advance in a storage means (not shown).

  The three-dimensional model image portion detecting means 9 uses the color (the three-dimensional model is formed in a single color in advance) for each frame of the live-action image signal photographed by the photographing camera 4 for the outer edge portion of the image portion on which the three-dimensional model is projected. The information relating to the outer edge of the video part is output to the replacement means 11 as a three-dimensional model video part detection signal. The three-dimensional model video partial detection signal is arrangement data indicating the position where the CG image is arranged in the actual space (in the live-action video). That is, the three-dimensional model video part detecting means 9 performs image processing on the real video signal to detect the video part of the three-dimensional model that is a detection target.

  The replacement unit 11 converts the CG image drawn (generated) by the rendering unit 7 into the real video signal photographed by the photographing camera 4 based on the three-dimensional model video part detection signal output by the three-dimensional model video part detection unit 9. , And output as a composite video signal.

  The composite video signal output from the replacement means 11 can be displayed on a display device (not shown), or can be input to a broadcast wave transmission device (not shown) and transmitted as a broadcast wave. Is possible.

  The three-dimensional model image portion detection means 9 and the replacement means 11 correspond to each processing means of a conventional chroma key device (not shown).

  According to this video composition device 1, the 3D CG data output control means 5 outputs the 3D CG data stored in the 3D CG data storage means 3 to the 3D model modeling device 2. At this time, the three-dimensional model forming apparatus 2 forms a three-dimensional model when the three-dimensional CG data is input. When the photographed camera 4 photographs a modeled solid model, a person, and the like, photographing information including the position and direction of the photographing camera 4 sensed by the sensor 6 and the zoom amount (zoom rate) is input. Then, the photographing means input from the sensor 6 by the rendering means 7, the placement information indicating the position where the pre-stored three-dimensional model is placed, and the three-dimensional CG input from the three-dimensional CG data storage means 3. A CG image (two-dimensional CG image) is drawn based on the data. Then, the three-dimensional model image portion detection means 9 detects the image portion in which the three-dimensional model in the live-action image signal photographed by the photographing camera 4 is projected as a three-dimensional model image portion detection signal, and the replacement means 11 The detected video portion is output as a composite video signal in which the CG image (two-dimensional CG image) drawn by the rendering unit 7 is replaced. For this reason, a person such as a performer appearing in a program does not need to wear sensors and special glasses that have been worn by the conventional mixed reality technology, and can operate while looking at the actually modeled three-dimensional model. . Further, since the three-dimensional model is replaced with a CG image, a synthesized video signal (synthesized video) in which a person behaves naturally with respect to the CG image can be output.

  In other words, according to this video composition device 1, a person can operate while looking at an actual three-dimensional model and pointing to the three-dimensional model. 7 can be replaced with the CG image drawn in step 7, so that a composite video signal (composite video) in which the person behaves naturally with respect to the CG image can be obtained.

(Operation of video composition device)
Next, the operation of the video composition apparatus 1 will be described with reference to the flowchart shown in FIG. 2 (see FIG. 1 as appropriate). In the explanation of this operation, before the photographing by the photographing camera 4 is started, the three-dimensional CG data is output from the three-dimensional CG data storage means 3 to the three-dimensional model shaping apparatus 2, and the three-dimensional model shaping apparatus 2 uses the three-dimensional model shaping apparatus 2. Is premised on modeling.

  First, a live-action image signal obtained by photographing a three-dimensional model and a person with the photographing camera 4 is input to the three-dimensional model image portion detecting unit 9 and the replacing unit 11 of the image synthesizing apparatus 1 (S1).

  In addition, the video composition device 1 receives (acquires) shooting information including the position, direction, and zoom amount (zoom rate) of the shooting camera 4 output from the sensor 6 (S2).

  Then, the video composition device 1 renders (draws) a CG image by the rendering unit 7 based on the photographing information, the arrangement information, and the three-dimensional CG data (S3).

  In addition, the video composition apparatus 1 uses the stereoscopic model video part detection unit 9 as a stereoscopic model video part detection signal to replace the video unit in which the stereoscopic model is projected from the actual video signal input from the photographing camera 4. Output (S4).

  Then, the video synthesizing apparatus 1 replaces the CG image drawn by the rendering unit 7 with the video portion on which the three-dimensional model is projected by the replacing unit 11, and outputs the resultant as a synthesized video signal (synthesized video) (S5).

(Example of composite video)
Next, an example of a synthesized video that is synthesized and output by the video synthesizer 1 will be described with reference to FIG. FIG. 3 shows an example of a synthesized video that is synthesized by the video synthesizer 1 and output.
As shown in FIG. 3, the person is pointing to the CG image while looking at the CG image of a building such as a high-rise building (actually while looking at the three-dimensional model).

  According to this, a person such as a performer appearing in a program or the like can operate by pointing to a specific part of the three-dimensional model while wearing a simple three-dimensional model without wearing a sensor or special glasses. Therefore, it is possible to perform a natural behavior (natural operation, line of sight) with respect to the CG image.

In addition, an example of how the synthesized video that is synthesized and output by the video synthesizing device 1 will be described with reference to FIG. 4 (see FIG. 1 as appropriate).
4 (a-1) to 4 (c-2) show a direction in which a person such as a performer appearing in a program stands around the stereoscopic model while pointing to the stereoscopic model (with the photographing camera 4). This shows a case where the shooting direction is changed.

  In FIG. 4 (a-1, a-2), FIG. 4 (a-1) has shown that the three-dimensional model r of "building" is located in the left direction toward a person. For example, when the photographing model 4 is used to photograph the three-dimensional model r from the direction in which the person is looking at the three-dimensional model r, the position, direction, and zoom amount of the photographing camera 4 when the three-dimensional model r is photographed by the photographing camera 4. A CG image is generated based on the shooting information including the position information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data used when forming the three-dimensional model r. Then, the CG image is combined with the actual video signal. FIG. 4A-2 shows the result, that is, that the synthesized video (synthesized video signal) is displayed.

  4 (b-1, b-2), FIG. 4 (b-1) shows that the three-dimensional model r of “building” is located in front of the person. For example, when the 3D model r is shot with the shooting camera 4 from the direction in which the person is looking at the 3D model r, the position of the shooting camera 4 when the 3D model r is shot with the shooting camera 4 from the direction of the person. A CG image is generated based on the photographing information including the direction and the zoom amount, the arrangement information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data used when the three-dimensional model r is formed. Then, the CG image is combined with the actual video signal. FIG. 4B-2 shows the result, that is, that a synthesized video image (synthesized video signal) is displayed.

  Further, in FIG. 4 (c-1, c-2), FIG. 4 (c-1) shows that the three-dimensional model r of “building” is located in the right direction toward the person. For example, when the photographing model 4 is used to photograph the three-dimensional model r from the direction in which the person is looking at the three-dimensional model r, the position, direction, and zoom amount of the photographing camera 4 when the three-dimensional model r is photographed by the photographing camera 4. A CG image is generated based on the shooting information including the position information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data used when forming the three-dimensional model r. Then, the CG image is combined with the actual video signal. FIG. 4C-2 shows the result, that is, that the synthesized video (synthesized video signal) is displayed.

  4D shows the positional relationship between the three-dimensional model r and the person, FIG. 4A shows the position of the person shown in FIG. 4A-1, and FIG. 4 (b-1) shows the position of the person, and (c) shows the position of the person shown in FIG. 4 (c-1). That is, when the position, direction, and zoom amount of the photographing camera 4 change from moment to moment, the generated CG image changes following the change.

  That is, in the video composition device 1, the rendering unit 7 sequentially changes the CG image to be drawn in accordance with the photographing information including the position, direction, and zoom amount of the photographing camera 4 input from the sensor 6, and the replacement is performed. The synthesized video signal synthesized by the means 11 can also be changed sequentially.

  As mentioned above, although embodiment of this invention was described, this invention is not limited to the said embodiment. For example, in the present embodiment, the video composition apparatus 1 has been described. However, a video composition program in which processing of each component of the video composition apparatus 1 is described in a general-purpose computer language (may be a special assembler language) It can be considered. In this case, the same effect as that of the video composition device 1 can be obtained.

1 is a block diagram of a video composition device according to the present invention. 3 is a flowchart for explaining the operation of the video composition apparatus shown in FIG. 1. It is a figure explaining the example of the synthetic | combination video output from the video synthesizing | combining apparatus shown in FIG. (A-1) shows a case where a person is located on the right side of the three-dimensional model, and (a-2) shows an example of a composite image in that case. (B-1) shows a case where a person is located in front of the three-dimensional model, and (b-2) is a diagram showing an example of a composite image in that case. (C-1) shows a case where a person is located on the left side of the three-dimensional model, and (c-2) shows an example of a composite image in that case. (D) is the figure which showed the positional relationship of a solid model and a person.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image composition apparatus 2 Three-dimensional model shaping apparatus (three-dimensional model shaping means)
3 Three-dimensional CG data storage means 4 Photographing camera (photographing means)
5 3D CG data output control means (3D CG data output means)
6 sensor (photographing information acquisition means)
7 Rendering means 9 Three-dimensional model image partial detection means 11 Replacement means A Image composition system r Three-dimensional model

Claims (4)

In the image where the 3D model is projected in the image of the 3D model modeled by the 3D model modeling means and the person who explains using the 3D model by the shooting means. , A video synthesizing device that replaces and outputs a CG image drawn according to the three-dimensional CG data,
3D CG data storage means for storing the 3D CG data;
Three-dimensional CG data output means for outputting the three-dimensional CG data stored in the three-dimensional CG data storage means to the three-dimensional model shaping means;
Based on photographing information including the position, direction and zoom amount of the photographing means, arrangement information indicating a position where the three-dimensional model is arranged, and three-dimensional CG data accumulated in the three-dimensional CG data accumulating means. Rendering means for rendering the CG image;
A three-dimensional model video part detecting means for detecting a video part in which the three-dimensional model is projected in the video;
A replacement means for replacing the CG image drawn by the rendering means with the video portion detected by the three-dimensional model video portion detection means;
A video synthesizing apparatus comprising:   2. The image synthesizing apparatus according to claim 1, wherein the three-dimensional model forming unit uses a principle of a thin film stacking method in rapid prototyping technology. In the image where the 3D model is projected in the image of the 3D model modeled by the 3D model modeling means and the person who explains using the 3D model by the shooting means. In order to replace and output the CG image drawn according to the three-dimensional CG data,
Three-dimensional CG data output means for outputting the three-dimensional CG data from the three-dimensional CG data storage means for storing the three-dimensional CG data to the three-dimensional model shaping means;
Rendering means for rendering the CG image based on shooting information including the position, direction and zoom amount of the shooting means, arrangement information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data;
3D model image part detecting means for detecting an image part of the 3D model projected in the image;
Replacing means for replacing the CG image drawn by the rendering means with the video portion detected by the three-dimensional model video portion detecting means;
A video composition program characterized by functioning as Replace the CG image drawn in accordance with the 3D CG data in the video image of the 3D model in the video of the 3D model and the person who will explain using the 3D model. An image composition system for output,
3D CG data storage means for storing the 3D CG data;
Three-dimensional CG data output means for outputting the three-dimensional CG data stored in the three-dimensional CG data storage means;
Three-dimensional model modeling means for modeling the three-dimensional model according to the three-dimensional CG data output by the three-dimensional CG data output means;
Shooting information acquisition means for acquiring shooting information including the position, direction and zoom amount of the shooting means;
Based on the shooting information acquired by the shooting information acquisition means, the arrangement information indicating the position where the three-dimensional model is arranged, and the three-dimensional CG data stored in the three-dimensional CG data storage means, Rendering means for rendering a CG image;
A three-dimensional model video part detecting means for detecting a video part in which the three-dimensional model is projected in the video;
A replacement means for replacing the CG image drawn by the rendering means with the video portion detected by the three-dimensional model video portion detection means;
A video synthesizing system comprising:

Images