JP2001320730A - Image composing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that color information used for the discrimination of a foreground image and a background image is deteriorated, accuracy of discrimination is lowered, and the quality of an obtained composite image is, as a result, lowered, when a device to generate a background image for performing an image composite by using a chroma-key is shared by a plurality of image productions and picture compression is performed to improve the use efficiency of a network. SOLUTION: A background of image information photographed by a camera 101 is substituted by a background color generated by a background color generator 103 based on key frame information generated by a background displace 104 from a color separator 102, compressed by an image compressor 105, and then transmitted by a transmission transmitter 106 and a transmission receiver 107 to a remote place. Transmitted compressed image information is expanded by an image expander 108 and then composed with a background image generated by a background image generator 110 by an image compositer 111 based on key frame information generated by a color separator 109.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image synthesizing apparatus generally called a chroma key, which synthesizes one image by synthesizing a foreground image and a background image.
The present invention relates to an image synthesizing device that synthesizes a foreground image transmitted via a network or the like and a background image.

[0002]

2. Description of the Related Art In video production, an image synthesizing method of fitting a foreground image of a person or the like into another background image is often used. In general, as a means for extracting a foreground image, a method called a chroma key, in which a foreground image is photographed in front of a background of a specific color such as blue and the difference in color information between the foreground image and the background image is used.

That is, as shown in FIG. 8, a person 801 to be a foreground image is photographed by a camera 803 in front of a background 802 colored in blue. The captured image information is input to a color separator 804, which detects a color of a background image to generate a key frame signal which is a signal for switching between a foreground image and a background image. Background image generator 805
Is a VTR that outputs a separately shot image, a computer graphics generator, or the like, and generates a background image. The synthesizer 806 receives the image from the camera 803 and the image from the background image generation device 805 as input, and synthesizes them using the key frame signal generated by the color separator 804.

[0004] As a result, the background colored in solid blue behind the person 801 is replaced with the background image generated by the background image generator 805, as if the person 801 is
An image as if in front of the background can be synthesized.

[0005]

As a recent trend, with the rapid development of computer technology, a computer graphics generator is often used as the above-mentioned background image generator. As the demand for background images generated by computer graphics becomes more sophisticated and computer graphics equipment becomes more expensive, advanced skills are required for computer graphics production. In some cases, it is difficult to prepare foreground images and background images on the spot where images are produced.

Therefore, in response to a request for synthesizing a plurality of images,
In order to supply a background image to each, the configuration of FIG. 9 is adopted. In the figure, components having the same numbers as those in FIG. 8 perform the same operations. Image information from the camera 803 is compressed by the image compressor 807 and transmitted to the network 809 by the transmission / transmission device 808. The transmission receiving device 810 receives the compressed image information from the network 809, and the image decompressor 811 restores the compression by the image compressor 807. The image information restored by the image decompressor 811 is processed in the same process as in FIG. 8 by the color separator 804, the background image generator 805, and the image synthesizer 8
In step 06, image composition is performed. The synthesized image is compressed again by the image compressor 812, and the transmission and transmission device 813,
The image is restored by the image decompression device 815 via the network 809 and the transmission / reception device 814.

With this configuration, one background image generation device can respond to a request for synthesizing a plurality of images via a network. In addition, as long as a place where a foreground image is captured, a place where a background image is generated, and a place where a final composite image is used are connected via a network, each operation can be performed at any place.

However, since the color information is used for switching between the foreground image and the background image as described above, if the color information is lost during the compression / decompression process, the switching between the background image and the foreground image becomes unstable, The quality of the finally obtained composite image is impaired. To eliminate this effect, if the image compressor and the image decompressor are omitted from the configuration of FIG.
9, the amount of information that must be transmitted increases, and if the capacity of the network is exceeded, transmission itself becomes impossible,
Even when the transmission is possible, the burden of the usage fee related to the transmission increases.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and provides an image synthesizing apparatus via a network that achieves both high network use efficiency by image compression and sufficient synthesized image quality. The purpose is to:

[0010]

According to the image synthesizing apparatus of the present invention, in the image information composed of the foreground image and the background of the specific color, the background color is changed to the background color generated by the background color generating means. A second key frame generating means for generating key frame information by comparing luminance information, hue and saturation of transmitted image information based on the background color generated by the background color generating means. It is.

According to the present invention, the background of the image information to be transmitted is made simple by the above-mentioned configuration, and not only the color information but also
Since the foreground image and the background are identified using the luminance information, the accuracy of identification on the receiving side can be improved, and the quality of the synthesized image can be improved.

Further, the image synthesizing apparatus of the present invention transmits the key frame information output by the key frame generating means together with the image information.

According to the present invention, since the key frame information is transmitted independently of the image information by the above-described configuration, the accuracy of identification is not impaired and the utilization efficiency of the network is improved.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first invention of the present invention is directed to a first image information composed of a foreground image and a background of a specific color.
First key frame generating means for generating key frame information for identifying the foreground image and the background based on the color information of the background, and a preset luminance, hue,
A background color generating means for generating a background color having a uniform color saturation, and the background of the first image information based on the key frame information outputted by the first key frame generating means. Replace with the background color generated by
Background replacement means for generating image information of the following, transmission means for compressing and transmitting the second image information, and generation of the background color generation means for the second image information transmitted by the transmission means. Second key frame generating means for generating key frame information for identifying the foreground image and the background color based on the color information of the background color to be generated, and a key frame output by the second key frame generating means. Based on the information,
An image synthesizing device comprising an image synthesizing unit for synthesizing the foreground image and a desired background image. With this configuration, the image synthesizing device of the present invention has The background is replaced with the background image generated by the background color generating means, and the second key frame generating means
The key frame information is generated by comparing the luminance information, hue, and saturation of the background color generated by the image information of the background and the background color generated by the background color generation means, so that the pixel information of the background can be simplified, and the foreground image and the background Since not only the color information but also the luminance information is used for the identification, the identification accuracy on the transmission side can be improved.

According to a second aspect of the present invention, in the first aspect, a foreground image is extracted from the first image information based on the key frame information generated by the first key frame generating means, and the foreground image of the foreground image is extracted. The foreground image analysis means for extracting statistical information of luminance, hue and saturation, and the background color generation means, based on the statistical information output by the foreground image analysis means, calculates the most from the average hue of the foreground image. An image synthesis method that generates a background color with a uniform hue, hue, and saturation that is a separated hue and has the least frequency of occurrence and the highest saturation among the brightnesses. With this configuration, the image synthesizing apparatus of the present invention generates a background color that can be easily distinguished from the foreground image based on the statistical information of the foreground image output from the foreground image analysis unit by the background color generation unit. It can improve the accuracy of the identification of the image and the background.

The background color replacing means always replaces the pixel information of the specific coordinates set in advance with the background color generated by the background color generating means. Can be used to recognize the background color generated by the background color generation means.

According to a third aspect of the present invention, for the image information composed of a foreground image and a background of a specific color, the image information for identifying the foreground image and the background based on the color information of the background is provided. Key frame generating means for generating key frame information;
Transmitting means for compressing and transmitting the key frame information output by the key frame generating means and the foreground image, and an image for synthesizing the foreground image and the background image based on the key frame information transmitted by the transmitting means And a synthesizing unit. With this configuration, the image synthesizing apparatus of the present invention transmits the key frame information independently of the image information. it can.

Further, there are provided contour information extracting means for extracting contour information of the foreground image from the key frame information, and key frame reproducing means for reproducing a key frame from the contour information. It can also be transmitted independently.

Hereinafter, an embodiment of the present invention will be described with reference to FIG.
This will be described with reference to FIG.

(Embodiment 1) FIG. 1 is a block diagram of an image synthesizing apparatus according to Embodiment 1 of the present invention. In FIG. 1, reference numeral 101 denotes a camera for capturing an image serving as a foreground image, and 102, information for identifying a foreground image and a background image by detecting and separating a color of a background image in image information captured by the camera 101. Is a color separator which is a first key frame generating means for generating image data, 103 is a background color generator for generating image information to be a background, and 104 is imaged by the camera 101 based on information from the color separator 102. A background replacement unit that replaces the background portion of the image information with the background color generated by the background color generator 103; 105, an image compressor that performs image compression on the image information output from the background replacement unit 104 in accordance with the DV system; 106, a transmission / transmission device for transmitting the compressed image information output from the image compressor to a network, 107
Is a transmission receiving device that receives the compressed image information transmitted from the transmission transmitting device 106 from the network, 108 is an image decompressor that decompresses the compressed image information conforming to the DV system received by the transmission receiving device 107, and 109 is an image decompressor. Vessel 108
A color separator 11 serving as second key frame generating means for outputting information for identifying a foreground image and a background image by detecting and separating a background color in the output image information;
0 is a background image generator for generating a background image, and 111 is a background image generated by the background image generator 110 by converting the background portion of the image information expanded by the image expander 108 based on the information from the color separator 109. The image synthesizer to be replaced. Here, the image compressor 106, the transmission transmission device 106, the transmission reception device 107, and the image decompression device 108 correspond to transmission means.

The operation of the image synthesizing apparatus configured as described above will be described below.

The camera 101 captures an image serving as a foreground image in front of a specific color background. A person is often photographed as the foreground image, and a color that is complementary to the skin color of the person, for example, blue is often selected as the background color. The color separator 102 receives image information captured by the camera 101, detects and separates the color of the background image in the image information, generates key frame information for identifying the foreground image and the background image, and performs background replacement. Output to the device 104. The background color generator 103 generates image information in which luminance, hue, and saturation as a background color are uniform over the entire screen. The background replacer 104
According to the key frame information output from the color separator 102,
For the foreground part of the image, the image information from the camera 101 is selected, and for the background part, the image information generated by the background color generator 103 is selected. Replace with the background color generated by. The image compressor 105 adds SMPTE (Society of Motion Pictu) to the image information output from the background replacement unit 104.
re and Television Engineers) DV (Di
Gital Video), and the transmission / transmission apparatus 106 transmits the compressed image information output from the image compressor 105 to a remote place via the network in combination with the transmission / reception apparatus 107.

An image decompressor 108 decompresses the compressed image information conforming to the DV system received by the transmission receiver 107 to the original image information. The color separator 109 detects and separates the background color generated by the background color generator 103 from the image information expanded by the image expander 108 to generate key frame information for identifying the foreground image and the background image. Image synthesizer 1
Output to 11 The background image generator 110 is a computer that generates a VTR or computer graphics.
Output to 1. The image synthesizer 111 selects the image information from the image expander 108 in the foreground portion of the image and the image information generated by the background image generator 110 in the background portion, based on the key frame information output from the color separator 109. The image information from the image decompressor 108 and the image information from the background image generator 110 are combined.

By the above processing, the background image generator 110
Can be synthesized as if there is a foreground image of a person or the like photographed by the camera 110 in front of the background generated by.

FIG. 2 is a schematic diagram showing combinations of pixels that are simultaneously compressed when the image compressor 105 performs image compression in accordance with the DV system. In the DV system, five distant pixel blocks on the screen are combined into one, and compressed. The five pixel blocks are coded, and are combined so that the combined code length becomes a fixed length. When the image information in the pixel block is complicated, the code length becomes long, and when the image information is simple, the code length becomes short. In the DV system, 5 of the pixel blocks to be collected are
The two locations are arranged so as to be dispersed on the screen as shown by a, b, c, d, and e in FIG. 2, and the code lengths assigned to the respective pixel blocks in the fixed length total code length , The code length is more effectively allocated than when the total code length is simply divided into five.

When image synthesis is performed, only the foreground portion of the entire image is finally used. Therefore, when the code length is preferentially assigned to the foreground part during compression, the definition of the foreground part is improved, and in particular, the boundary between the background image and the foreground image is sharpened. The separation accuracy between the image and the background image is improved. As shown in the conventional example, when image information from a camera is compressed as it is, the image of the background part is an image of a specific color, but due to shooting with the camera, the luminance is determined by the degree of illumination and the characteristics of the lens. Therefore, the code length is also consumed in pixel blocks such as b, c, d, and e in FIG. 2 where it is not desired to assign a code length because the noise is inevitably mixed. The amount of codes to be assigned to the pixel blocks in FIG. This causes the quality of the final synthesized image to deteriorate.

In the image synthesizing apparatus of the present invention shown in FIG. 1, the background portion is generated by the background color generator 103 by the background replacement unit 104 prior to the image compression.
The background image information having a constant saturation is replaced. As a result, the image information of the pixel block in the background portion becomes the simplest state, so that the code length assigned to the background portion is minimized, and a sufficient code length can be assigned to the foreground portion. As a result, the separation accuracy of the foreground image and the background image in the color separator 109 at the transmission destination is improved, and the quality of the synthesized image obtained from the image synthesizer 111 can be improved.

(Embodiment 2) FIG. 3 shows an image decompressor 1 input to a destination color separator 109 in Embodiment 1.
FIG. 18 is a signal waveform diagram in which a boundary portion between a foreground image and a background image of image information output by reference numeral 08 is enlarged. In the image compression of the DV system, information is transmitted at a ratio of one piece of color information to four pieces of luminance information per unit time. Therefore, at the boundary between the foreground image and the background image, compared to the luminance indicated by the curve in FIG.
The color information changes gradually as shown by the curve in FIG. In FIG. 3, the time from t0 to t3 is the foreground image and the background image is from time t4. However, since the color changes gradually, the color information of the pixel at t3 is almost close to the background, and the foreground image is only color information. When the background image is identified, the pixel at t3 is erroneously identified as the background. Further, when the background is still photographed by the camera as in the conventional example, the luminance of the background is not constant as described above, so that information on the luminance cannot be directly used for discriminating the foreground image from the background image. .

In the image synthesizing apparatus according to the present embodiment, the background portion is replaced by the background color generator 103 with background image information having a constant luminance, hue and saturation before the image compression. Therefore, the luminance of the background image can be uniquely determined.

The color separator 109 first compares luminance information before comparing color information. That is, in the example of FIG. 3, the pixel at t3 is determined to be a background image in the color information, but does not match the luminance of the background image in the luminance information. Is identified.

As described above, since the luminance information can be used for discriminating the foreground image from the background image, the accuracy of the discrimination can be improved, and the quality of the synthesized image obtained from the image synthesizer 111 can be improved. Can be.

(Embodiment 3) In this embodiment, a video synthesizing apparatus for generating a background color for replacing a background portion of image information captured by a camera using statistical information of a foreground image will be described.

FIG. 4 is a block diagram of an image synthesizing apparatus according to Embodiment 3 of the present invention. In FIG. 4, the same reference numerals are given to components performing the same operation as in the first embodiment, and description thereof will be omitted.

In FIG. 4, reference numeral 112 denotes a foreground image analyzer which extracts a foreground image from image information from the camera 101 from key frame information from the color separator 102 and calculates statistical information of the foreground image, and 113 denotes a foreground image analyzer. A background color generator 112 generates a background color from the calculated statistical information.

The difference between the present embodiment and the first embodiment is as follows.
The foreground image analyzer 112 is provided in the present embodiment because the background color generator 113 generates the background color based on the statistical information calculated by the foreground image analyzer 112. The operation of the color generator 113 will be mainly described.

The foreground image analyzer 112 receives the key frame information from the color separator 102, extracts the foreground image from the image information from the camera 101, and obtains the average value of the hue of the foreground image and the luminance as statistical information of the foreground image. Calculate the frequency of occurrence of the value. The background color generator 113 includes a foreground image analyzer 112.
From the statistical information of the foreground image output from the foreground image, from the hue that is the farthest from the average hue of the foreground image, has the least frequently occurring luminance value, and select the background color with the highest saturation in the luminance value. Generated and output to the background replacement unit 104.

In this embodiment, the background color generator 113
However, since the background replacement unit 104 can generate a background color that is most easily distinguished from a foreground image as a background color to be replaced with a background image of image information output from the camera 101,
It becomes easier to distinguish between the foreground image and the background image in the color separator 109 at the transmission destination. As a result, the probability of occurrence of erroneous identification in the color separator 109 can be reduced, and the image synthesizer 1
11 can be improved.

(Embodiment 4) In Embodiment 3, the replaced background color changes depending on the statistical characteristics of the foreground image. In the case of an image synthesizing apparatus in which the background color is determined in advance by the transmission source and the transmission destination, the color separator 109 needs to recognize the background color before generating a key frame.

In this embodiment, the background replacement unit 104 always selects the background image from the background color generator 113 at a specific coordinate, for example, at the upper left corner of the screen. The color separator 109 at the transmission destination recognizes the background color generated by the background color generator 113 from the information of the pixel at the specific coordinates, and uses it for discriminating the foreground image from the background image.

In this embodiment, the background replacement unit 10
4 describes the specific coordinates as the upper left corner of the screen, but is not limited to this.

(Embodiment 5) In the conventional example, color information is used to distinguish between a background image and a foreground image, and image compression is used in transmitting the color information. The image synthesizing apparatus according to the present embodiment will be described as an image synthesizing apparatus that transmits key frame information independently of image information.

FIG. 5 is a block diagram of an image synthesizing apparatus according to Embodiment 5 of the present invention.

In FIG. 5, reference numeral 101 denotes a camera for photographing an image to be a foreground image in front of a background of a specific color, and 102 detects and separates the color of the background image in the image information photographed by the camera 101. A color separator 105 for generating information for identifying a foreground image and a background image; 105, an image compressor for performing image compression based on the DV system on image information captured by the camera 101; 1
05, a transmission / transmission device for transmitting the compressed image information output from the key frame compressor 114 to the network,
Reference numerals 7 and 116 denote transmission receivers for receiving the compressed image information transmitted from the transmission transmitter 106 and the transmission transmitter 115 from the network, respectively.
Is an image decompressor that decompresses the received compressed image information conforming to the DV system, 110 is a background image generator that generates a background image, and 111 is an image decompressor that generates a background image from the image decompressor 108 based on key frame information output from the key frame decompressor 117 An image synthesizer for synthesizing the output image information and the background image generated by the background image generator 110, a key frame compressor 114 for compressing a key frame image, and a key frame 117 for converting the compressed key frame information to an original key frame. It is a key frame decompressor that decompresses information. Here, the transmission transmitting devices 106, 1
15. The transmission receivers 107 and 116 correspond to transmission means.

The operation of the image synthesizing apparatus configured as described above will be described below.

The camera 101 captures an image to be a foreground image in front of a background of a specific color, and converts the captured image information into a color separator 1.
02, output to the image compressor 105. Color separator 102
Receives the image information from the camera 101 and detects and separates the color of the background in the image information to generate key frame information for identifying the foreground and the background. The key frame compressor 102 performs image compression in accordance with the DV system on the key frame information generated by the color separator 102, for example, image compression as a binary image in which the foreground is white and the background is black. The transmission transmission device 115 and the transmission reception device 116 transmit the compressed key frame information compressed by the key frame compressor 114 to a remote place via a network. The key frame expander 117 expands the compressed key frame information conforming to the DV system to the original key frame information.

On the other hand, the image compressor 105 compresses image information from the camera 101. The transmission transmitting device 106 and the transmission receiving device 107 form a pair and transmit the image information compressed by the image compressor 105 to a remote place via a network. The image decompressor 108 is a transmission receiving device 10
7 expands the compressed image information output to the original image information.

The background image generator 110 is a VTR or a computer that generates computer graphics, and reproduces or generates a background color. Image synthesizer 111
Selects the image information from the image expander 108 in the foreground portion of the image and the image information from the background image generator 110 in the background portion, based on the key frame information output from the key frame expander 117. Thus, an image can be synthesized as if a foreground image such as a person photographed by the camera 101 exists in front of the background generated by the background image generator 110.

In this embodiment, by providing an independent transmission means for transmitting the key frame information, the identification accuracy is not impaired. In addition, since the compression of the DV system can reduce the compression by about 1/5, the use efficiency of the network is higher than the case where the transmission is performed independently and the compression is not used together.

Although the key frame compressor 114 compresses the key frame information by the DV system, the present invention is not limited to this. For example, Huffman coding is used as a compression method of the key frame compressor 114. Also, the key frame decompressor 117 to be paired performs decompression based on Huffman coding. The area occupied by the foreground image on the screen is
It is a continuous area and the correlation between lines is high. Huffman coding shows high efficiency in compressing this type of information.

(Embodiment 6) In the first embodiment, the key frame information is compressed and transmitted. In the present embodiment, the key frame information is transmitted using a chain of vectors expressing contour information. The image synthesizing device will be described.

FIG. 6 is a block diagram of an image synthesizing apparatus according to Embodiment 6 of the present invention. In FIG. 6, the same reference numerals are given to components performing the same operation as in the fifth embodiment, and description thereof will be omitted.

In FIG. 6, reference numeral 118 denotes a contour information extractor for extracting contour information from the key frame information output from the color separator 102, and 119 denotes a transmission / transmission device for sending the contour information output from the contour information extractor 118 to a network. , 1
Reference numeral 20 denotes a transmission receiving device that receives the contour information transmitted from the transmission transmitting device 119 from the network, and reference numeral 121 denotes a key frame regenerator that restores the original key frame information from the contour information received by the transmission receiving device. Here, the transmission transmission device 119 and the transmission reception device 120 correspond to transmission means.

The difference between this embodiment and the fifth embodiment is that
Since this is a method of transmitting key frame information, in the present embodiment, the operation of the outline information 118 and the operation of the key frame reproducer 121 will be mainly described.

The contour information extractor 118 detects a boundary between the foreground and the background from the key frame information output from the color separator 102, and utilizes the fact that the boundary becomes a closed curve to convert the closed curve forming the boundary into a closed curve as shown in FIG. Is expressed as a chain of vectors in eight directions. Here, the chain of vectors starts at a certain coordinate on the closed curve, and the position of the next coordinate can be represented by one of the vectors in the eight directions shown in FIG.
If this processing is sequentially performed on all points of the flat curve shape, the closed curve can be represented by a chain of vectors.

The transmission transmitter 119 is the transmission receiver 120
And transmits vector information indicating a boundary of a key frame to a remote place via a network. The key frame reproducer 121 restores the vector information closed curve representing the boundary of the key frame, and restores the original key frame.

In this embodiment, a chain of vectors expressing contour information is used for transmitting key frame information. Since the vectors used here have only eight directions as shown in FIG. 7 and only three bits of information are required to distinguish them from each other, the transmission efficiency of key frame information can be made extremely high. .

In the present invention, the transmission transmitting apparatus and the transmission receiving apparatus transmit image information via a network. However, the present invention is not limited to this, and any transmission means capable of transmitting image information at a remote location may be used. .

In the present invention, the camera is used as an input means for image information composed of a foreground image and a background of a specific color. However, the present invention is not limited to this. VT recording recorded image information
The input may be made by reproducing R.

In the present invention, the image compressor is a DV
The image is compressed and decompressed by a method conforming to the method, but is not limited to this.

[0060]

As is apparent from the above description, according to the present invention, a plurality of apparatuses for synthesizing background images which can perform image synthesis via a network or the like and which are expensive and require a dedicated operator are provided. It is possible to provide an image synthesizing apparatus via a network that achieves both high network utilization efficiency by image compression and sufficient synthesized image quality.

[Brief description of the drawings]

FIG. 1 is a block diagram of an image synthesizing apparatus according to Embodiment 1 of the present invention.

FIG. 2 is an explanatory diagram of image compression of a DV system.

FIG. 3 is a waveform diagram of a boundary portion between a foreground image and a background image.

FIG. 4 is a block diagram of an image synthesizing apparatus according to Embodiment 3 of the present invention.

FIG. 5 is a block diagram of an image synthesizing apparatus according to a fifth embodiment of the present invention.

FIG. 6 is a block diagram of an image synthesizing apparatus according to a sixth embodiment of the present invention.

FIG. 7 is an explanatory diagram of a vector expressing a boundary between a foreground and a background.

FIG. 8 is a block diagram of a conventional image synthesizing apparatus.

FIG. 9 is a block diagram of a conventional image synthesizing apparatus via a network.

[Explanation of symbols]

 101 Camera 102 Color Separator 103, 113 Background Color Generator 104 Background Replacer 105 Image Compressor 106, 115, 119 Transmission Transmitter 107, 116, 120 Transmission Receiver 108 Image Decompressor 109 Color Separator 110 Background Image Generator Reference Signs List 111 Image synthesizer 112 Foreground image analyzer 114 Key frame compressor 117 Key frame decompressor 118 Outline information extractor 121 Key frame reproducer

Continued on front page F-term (reference) 5B057 AA20 BA02 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC03 CE08 CE17 CG01 DA08 DB02 DB06 DB09 DC16 DC25 5C066 AA12 BA17 CA05 CA11 DC01 DC06 ED00 ED02 ED03 EE01 EA11 GA01 GA01 GA03 AA12 AA13 AA26 BA06 CA12 5C079 HA01 HB06 LA02 LA07 LA26 LA40 LB13 MA02 NA13 5L096 AA02 AA06 BA20 CA07 DA01 FA15

Claims (6)

[Claims] 1. Key frame information for identifying the foreground image and the background based on the color information of the background is provided for first image information composed of a foreground image and a background of a specific color. First key frame generating means for generating; background color generating means for generating a background color having uniform luminance, hue, and saturation preset; and key frame information output by the first key frame generating means A background replacement unit that generates the second image information by replacing the background of the first image information with a background color generated by the background color generation unit; and compresses the second image information. Transmitting means for transmitting the foreground image and the background color with respect to the second image information transmitted by the transmitting means based on the color information of the background color generated by the background color generating means Keyframe information for A second key frame generating means for generating; and an image synthesizing means for synthesizing the foreground image and a desired background image based on the key frame information output by the second key frame generating means. Image synthesizing device. A second key frame generation unit that compares the second image information transmitted by the transmission unit with luminance information, hue, and saturation of a background color generated by the background color generation unit and performs key frame generation; The image synthesizing apparatus according to claim 1, wherein the information is generated. 3. A foreground image is extracted from first image information based on key frame information generated by a first key frame generation unit, and statistical information of luminance, hue, and saturation of the foreground image is extracted. A foreground image analyzing unit, wherein the background color generating unit is a hue farthest from an average hue of the foreground image based on the statistical information output by the foreground image analyzing unit, and has a lowest frequency of occurrence. Luminance,
2. The image synthesizing apparatus according to claim 1, wherein a background color having a uniform luminance, hue, and saturation, which is the highest saturation among the luminances, is generated. 4. The background replacement means always replaces pixel information of a specific coordinate set in advance with a background color output by the background color generation means, and the second key frame generation means converts the pixel information of the specific coordinate from the pixel information of the specific coordinate. 4. The image synthesizing apparatus according to claim 3, wherein the background color generating means detects the luminance, hue, and saturation of the background color generated. 5. A key for generating key frame information for identifying the foreground image and the background based on the color information of the background, for image information composed of a foreground image and a background of a specific color. A frame generation unit, a transmission unit that compresses and transmits the key frame information and the image information output by the key frame generation unit, and, based on the key frame information transmitted by the transmission unit, the foreground image and a desired An image synthesizing apparatus, comprising: image synthesizing means for synthesizing a background image. 6. A key for generating key frame information for identifying the foreground image and the background based on the color information of the background, for image information composed of a foreground image and a background of a specific color. Frame generating means; contour information extracting means for extracting contour information of the foreground image from key frame information output by the key frame generating means; transmitting means for transmitting the contour information and the image information; Key frame reproducing means for restoring key frame information from the transmitted contour information, and image synthesizing means for synthesizing the foreground image and a desired background image based on the key frame information output by the key frame reproducing means. An image synthesizing apparatus characterized by the above-mentioned.