JPH01870A - television special effects equipment - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to television special effects equipment.

[Prior Art] A television image is a two-dimensional representation of a scene constructed by a television program producer. This scene may be a photograph of a real object or may be synthesized by artificial means (for example, a television graphics system), so the signal source of the video signal representing the scene is
Not only cameras or film scanners, but also frames and
It may be a buffer and a computer that adjusts the contents of the frame buffer. Typically, a scene is composed of two scene components, a foreground scene and a background scene, and the traveling matte technique is used to combine the scene components6.
For example, as shown in Figures 3a and 3b, if the foreground scene is a ring with a plain color matte background and the background scene is a rectangle with a contrasting color background, the foreground and background scenes can be combined. If so, the composite image will be as shown in Figure 3c.

The video conversion system is a system that manipulates video signals representing a scene, but it may also be used for special scenes, such as:
You can also move the scene to the right. A foreground video signal representing the scene of Figure 3a is fed to a conversion system, which transforms this video signal such that the ring of the scene of Figure 3a is shifted to the right, and the transformed video signal represents the scene of Figure 3d. If displayed, the video signal obtained by combining the transformed foreground signal with the background signal will represent the image shown in Figure 3e.

There are two main types of most conversion systems: forward conversion systems and frame-based reverse conversion systems. FIG. 4 shows a reverse conversion system based on known principles.Although the system of FIG. 4 does not appear to have previously existed, it will now be described as it is useful for understanding the present invention.

The system shown in FIG.
digitizes the input video signal under the control of the
The input video signal, which is written to the video frame buffer 12 by the address generated by the forward address generator 14, is an analog composite video signal in a conventional interlaced format, divided into its components (usually a luminance component and a chrominance component). chromaticity) component) and digitize each of these components. The sampling rate of each chrominance component is only half the sampling rate of the luminance component.

Frame buffer 12 is comprised of a memory that stores luminance components and a memory that stores chrominance components. However, there is no need to consider these components separately since they are processed in the same way within the conversion system.

Digitization of the image image effectively allows each scan line of the image to be divided into a small but finite number of pixels within a region (eg, 720 pixels).

The pixel position of the scene can be determined by the two-coordinate display address (U, V) of the input scene (e.g. FIG. 3a) and the forward address generator (writing means) 14.
The address space of the video frame buffer is systematized so that there is a one-to-one correspondence between the memory address where the video frame buffer is generated and the memory address where the video frame buffer is generated. Therefore, the display address (U.

Frame buffer 12 has three field memories, in which one field memory is written and simultaneously written to the other two.
Reading is performed from one memory.

To read the output video signal from the frame buffer 12, the read address counter 16 operates under the control of the read tarlock generator 17 to sequentially determine the position in the output scene (Figure 3d) of the addressed pixel. A series of addresses to be determined (X.

Y) is generated. The number of digits of the coordinate values X and Y is the same as the coordinate values U and ■, respectively. Therefore, the display address (U, V
) defines the position in the input display space, so the display address (X, Y) defines the same pixel location in the output display space. However, the display address (X, Y) is not directly used to read the output video signal from the frame buffer. A reverse address generator (address generation means) 18 receives the output scene display addresses (X, Y) and multiplies these addresses by the translation matrix T' to produce the corresponding memory address (X', Y'). occurs. Using this memory address, the user interface (I/
F) 19 supplies the translation matrix T' to the reverse address generator 18 to define the translation characteristics by the reverse translation system. For example, when performing a transformation that moves the input scene diagonally upward and to the left by an amount equal to the inter-pixel pitch, the transformation matrix
memory address <X', Y' generated in response to Y)
) is <X+1. Y+1). It is assumed here that the origin of the coordinate system is the upper left corner of the input and output scenes, and that the X and Y values increase toward the right and down, respectively.

In the general case, the memory address coordinates X' and Y' are smaller than the display address coordinates also has a large number of significant digits.

The reverse address is provided not only to frame buffer 12 but also to video interpolator 20. For each reverse address (X', Y'), frame buffer 1
2 outputs each digital word, but this digital
A word represents a xel array when the reverse address <x',y') surrounds the defined position. For example, if the address (
This digital word (data) represents the four pixels closest to the point defined by Combine the digital words into a single digital output word according to the fractional portion of the address (X', Y'). For example, when using the decimal system, the lowest digit of each coordinate X and Y is 1, but the coordinate
’ and Y’ are 1/10, and the counter 1
6 generates a read address (23, 6), and multiplication with the translation matrix T' changes this read address to the reverse address '(56, 3, 19, 8).
Frame buffer 12 uses this reverse address (5
6, 3, 19, 8), the address (56, 19
), (56,20), (57,19) and (57,20
), and interpolator 20 combines these digital words into a single digital output word with a weighting of 3:7 horizontally and 8:2 vertically.

This digital word defines the value to be generated at the output screen location defined by the display address (23.6).

Because the possible range of reverse addresses is wider than the range of memory addresses that define locations within frame buffer 12, a validly generated reverse address will address a location that does not exist within the frame buffer's address space. may be defined. Therefore, a reverse address is provided to the address boundary detector 22. Detector 22 generates a signal in response to an invalid reverse address (an address that defines a location outside the address space of frame buffer 12) that causes video blanking circuit 24 to It is also prohibited to output an output signal.

A key frame buffer 26, a key interpolator 28, in parallel with the video channel consisting of a video frame buffer 12, a video interpolator 20, and a video blanking circuit 24.
and a key blanking circuit 30.

The key signal provided to this key channel provides opacity information regarding the foreground video signal provided to the video channel. This opacity information defines the visible range of the background scene represented by the background video signal in a composite image (FIG. 3C) generated by mixing the foreground video signal and the background video signal under the control of the key signal. Outside the boundaries of the foreground object, the foreground scene is transparent (key=O) and the background scene is visible without modification by the foreground scene. If the foreground object is fully opaque (key=1), the background scene is completely obscured by the foreground object. However, if the foreground object is slightly transparent (0 x key x 1), the background video signal is mixed with the foreground video signal in proportion to the value of the key signal. As the video channel transforms the foreground scene, the keys must also be transformed to maintain foreground scene and key correspondence. Therefore,
The key signal is processed in the key channel in the same way that the foreground signal is processed in the video channel.

This key signal therefore undergoes the same two-dimensional transformation and interpolation as the foreground signal, and also undergoes the same address boundary blanking.

The transformation matrix T' is the mathematical reciprocal of the desired two-dimensional transformation T, so that reverse transformation systems are known as described above.

When a television picture consists of a foreground scene and a background scene, special effects can be used to make the screen more realistic and to make the picture consist of two (or more) independent scenes. Make sure it doesn't look like you're there.

An effect that can give a sense of reality to the screen is the addition of shadows. As shown in FIG. 5a, the foreground is a vertical column (or rectangular plate) 40, and the background is a uniformly colored vertical surface 43. Since the scene of FIG. 3 consists of two independent scenes, the shadow of pillar 40 is not projected onto surface 43. However, it is possible to change the image of a portion of the background scene to make it appear that a shadow 42 is present (Figure 5b).

Such effects can be achieved using conventional video special effects equipment.

DigiMole video mounting equipment is generally two-dimensional (or two-dimensional).
It is possible to perform signal processing (dimensionally) and create a composite scene (object and its shadow). First, a desired background scene is recorded on videotape. To simulate shadows, a key signal is input into a special effects device and a shadow matte (usually full screen or black) is applied to the video input of the special effects device. Note that the above-mentioned key signal is a signal that specifies a background portion where a shadow is formed by an object in the foreground. The output of the special effects device is a signal representing a black area with the same size and shape as the foreground object, but with a displacement ( offset).

This shadow part is recorded superimposed on the background. The foreground scene signal is then input into a special effects device in place of the shadow matte signal, and the foreground scene is processed two-dimensionally in the same manner as the shadow matte signal, but no displacement of the foreground scene is performed.

The converted foreground scene is recorded over a composite scene of the background and shadow areas. Alternatively to the method described above, if two special effects devices are available, the shadow matte signal and the foreground scene signal are input to separate special effects devices, and the key signal is applied to both special effects devices. One special effects device processes (transforms) the foreground scene signal, and the other special effects device creates a displaced shadow signal. The outputs of both devices are combined and
It is recorded over the background scene signal.

[Problems to be solved by the invention] In conventional television video special effects devices, in order to create a signal for the shadow of a two-dimensionally transformed object, signal processing is performed twice within one special effects device. There is a need. On the other hand, 2
If two special effects devices are available, the two signal processes can be performed simultaneously in separate devices, but two signal processes are still being performed.

Therefore, it is an object of the present invention to
Shadows of objects in the foreground scene can be created in the background scene.

SUMMARY OF THE INVENTION According to a preferred embodiment of the present invention, an input video signal representing a foreground scene and a key signal representing opaque portions of the foreground scene are processed. Two-dimensional conversion processing is performed on the input video signal to create a converted video signal.

The input key signal includes a series of digital data words that are read into a key frame buffer. The first consecutive address word (X
, Y) and convert this consecutive first address word to produce a consecutive second address word <X', Y'
) occurs.

A consecutive second address word is used to read the digital data word from the key frame buffer to generate a first converted key signal. Conversion of successive first address words is performed as follows.

That is, just as the video signal is two-dimensionally transformed in relation to the input video signal, the key signal is two-dimensionally transformed in relation to the input key signal. The consecutive second address words are transformed to create a consecutive third address word. This third address word is used to read the digital data word from the key frame buffer to generate a second translation key signal. The converted video signal is then combined with the first and second converted key signals.

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

The television special effects apparatus shown in FIG. 1 is similar to the apparatus shown in FIG. 4, except that it includes a reverse address generator 18, a shadow processing circuit 48, and the like. The output of the address boundary detector 22 is not provided to the key signal blanking circuit 30, and the second address boundary detector 46 is not provided to the key signal blanking circuit 30.

It is provided between the address generator and the key signal blanking circuit. Additionally, in the circuit of FIG. 4, a shadow processing circuit 48 is provided between the blanking circuit 24 and the background mixer 32.

Shadow address generator 44 receives a series of reverse address words output from reverse address generator 18 and temporarily stores each address word in reverse address register 50. Shadow address generator 44 also receives and stores a shadow offset word from user interface 19 in shadow offset register 52 . A shadow offset word represents the two-dimensional relationship between a foreground object and its shadow. For example, if the object in the foreground is a vertical column, the background is a vertical wall, and the light source is in the upper left corner, the column shape will be at the bottom right of the column. In this case, the shadow offset word represents displacement upward and to the left. Shadow address generator 44 further includes an adder 56 and a multiplexer 58. Adder 56 adds the words stored in registers 50 and 52. On the other hand, multiplexer 58
is the output of adder 56 and reverse address generator 18
The outputs of the two outputs are alternately selected, and an address signal in which the two outputs are multiplexed (alternately arranged in time) is output.

The multiplexed address signal described above is applied to a key frame buffer 26 and an interpolator 28, and a key signal blanking circuit rp&30 is provided with a multiplexed key signal. The second address boundary detector 22 receives the multiplexed address signals, and applies an inhibit signal to the blanking circuit 30 if both address signals are not valid signals. The multiplexed key signal consists of a key translation signal identical to the translation key signal provided by the key channel of the reverse translation device of FIG. 2, and a shadow key signal. The transformation key signal identifies opaque portions of the two-dimensionally transformed foreground scene. In the example briefly shown in Figure 5a, the foreground scene is an opaque column, and everything else is the background scene. The key signal has a value of 1 inside the boundary of the foreground column and a value of 0 in the pond area. On the other hand, the conversion key signal has a value of 1 within the boundary of the two-dimensionally converted column, and has a value of O in other parts (FIG. 5b). The conversion key signal has a value of 1 in the vertically lined portion (inside the boundary line), and the conversion key signal has a value of 0 in the horizontally lined portion.

The shadow key signal is identical to the transform key signal except that the foreground pillars are displaced according to the shadow offset word. In the case described above in connection with the generation of multiplexed address signals, the key identified by the shadow key signal takes the form as shown in Figure 5c. While the shadow offset word represents an upward and leftward displacement, the key signal specified by the inverse transformation by the transformer results in a downward and rightward displaced shadow key signal. As can be seen in Figure 5C, the key identifies the outline of the simulated shadow projected onto the vertical wall, and the 9
It is rotated 0 degrees. In this way, the outline of the shadow is
When the foreground video signal is converted, the foreground rJrlArl
pillar).

Therefore, the shadow outline will be located below and to the left of the transformed foreground object.

Shadow processing circuit 48 receives the video signal from the video channel and the multiplexed key signal from the key channel. The demultiplexer/interpolator 60 includes a key
It receives a multiplexed key signal from a channel, separates the multiplexed key signal and performs temporal interpolation, outputs a converted key signal from an output terminal 62, and outputs a shadow key signal from an output terminal 64. The foreground video signal is multiplied by the transform key signal in multiplier 66. The purpose of this multiplication is to create a foreground object signal in which unwanted parts of the foreground scene are masked out. Outside the boundaries of the transformed column, the signal of the foreground object is zero. The shadow key signal is applied to a multiplier 68 where the shadow key signal is
Multiplied by a coefficient that varies depending on the desired attenuation of the background video signal.

The attenuated shadow key signal is then further multiplied with the complement of the transformed key signal in multiplier 70. Therefore, the shadow key signal will be 0 for all points in the output scene where the transform key signal is 1 (Figure 5d).
The shadow key signal, which is the output of the 0 multiplier 70, is the fifth
It is added with the converted key signal in an adder 72 to obtain the output key signal shown in FIG. The shadow key signal output from multiplier 70 is further multiplied by a shadow matte signal in another multiplier 74. The shadow matte signal is a single color (the color is determined by user interface 19).
It is a full field signal representing the color (usually black) determined by the signal from the The output signal of multiplier 74 is a composite video signal that is an image signal having a shape specified by the shadow key signal from multiplier 70 and a color determined by the shadow matte signal. . Adder 76 adds the composite video signal from multiplier 74 and the foreground object signal from multiplier 66.

The output signal of multiplier 76 represents the transformed foreground 01 object image and the shadow. The shadow represented by the output signal of 0 multiplier 76 has the correct shape and is added to multiplier 68. It has an opacity determined by an opacity coefficient and a hue determined by a shadow matte signal. The foreground video signal output from the adder 76 consists of two video signals (
(each of which is multiplied by a corresponding key signal), so it is a shaped video signal.

The foreground video signal and key signal output from the shadow processing circuit 48 are supplied to the background mixer 32. Note that a background video signal is also added to the background mixer 32, as shown in the figure. Background mixer 32 combines the foreground and background scenes according to the value of the key signal and generates a video output.

Although the present invention has been described above based on the embodiments, it is obvious that the present invention is not limited to the embodiments described above, and modifications and changes to the embodiments described above will be obvious to those skilled in the art. for example,
In the above explanation, to simplify the explanation, the key signal is 2.
Generally, the key signal is 2 values.
It is not a value signal, but has levels between 0 and 1, for example 1024. By using a key signal that can take a value between 0 and 1, it is possible to make the foreground part semi-transparent.

Furthermore, the present invention provides 1. ! By writing the key signals into respective key frame buffers using two consecutive digital address words, the two key signals are By using one continuous address word to read digital words from the frame buffer, a foreground translation device can be implemented. Moreover, the invention is not limited to use with frame-based conversion devices, but can also be applied to field-based conversion devices.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram for explaining the embodiment of FIG. 1, and FIGS. 3a to 3e.
4 is a block diagram of a reverse conversion device, and FIG. 5 is a diagram for explaining the present invention in detail. 44...Shadow address generator 48...Shadow processing circuit

Claims (1)

[Scope of Claims] A television special effects device that processes an input video signal representing an image and an input key signal representing an opaque portion of the image and including a digital word, the apparatus comprising: a two-dimensional conversion into the video input signal; means for processing and converting the video input signal; a key frame buffer; and means for converting the digital data word into the key frame buffer;
means for writing to a buffer; means for generating a first address word; and means for converting said first address word to generate a second address word; and means for converting said first address word to generate a second address word; means for generating a third address word; and means for reading the digital data word from the key frame buffer using the second and third address words to generate first and second converted key signals, respectively. means for generating a key signal in a two-dimensional manner in relation to the input key signal, in the same manner as in converting the video signal two-dimensionally in relation to the input video signal. A television special effects device that converts an address word into a third consecutive address word.