JPH07184182A - Video conference equipment - Google Patents

Abstract

PURPOSE:To always make timing of video and audio signals to match each other with respect to a change in picture quality and the picture by varying a delay quantity of the audio signal matching a delay quantity of the video signal. CONSTITUTION:A de-frame arithmetic section 18 calculates a de-frame multi- value processing based on a TR obtained from a variable length decoding section 4, a mean value arithmetic section 19 calculates a mean value of de-frame quantity for n-frames, a delay arithmetic section 20 calculates a delay based on the mean value of the de-frame quantity and sets the delay to a counter section 12. Thus, a delay time of an audio signal by an audio signal delay section 11 is changed by a delay set to the counter section 12 to increase a delay time of the audio signal when the de-frame quantity is much, that is, the delay in the received video signal is large and to decrease a delay time of the audio signal when the de-frame quantity is less, that is, the delay in the received video signal is small conversely, then the delay time of the received audio signal is changed matching the delay in the received video signal and the output timing of the video signal is always matched with the output timing of the audio signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio signal delay method for adjusting the output timing of an audio signal to a video signal in a video conference apparatus for transmitting and receiving video / audio signals of a video conference, and image processing in the video conference apparatus. Regarding

[0002]

2. Description of the Related Art Conventionally, a video conference video / audio receiving apparatus of this type adjusts the output timing of the audio signal to the video signal which takes a long time for the encoding and the transmission at the time of the transmission when receiving the video signal and the audio signal. It has been done to delay audio signals.

FIG. 8 is a block diagram showing an example of a conventional video conference apparatus. The data separation unit 1 inputs the video signal and the audio signal multiplexed and sent from the transmission line, separates these two signals, and the video signal to the error correction unit,
The audio signal is output to the audio delay unit 11. The error correction unit 2 corrects the error of the input video signal, sends it to the input buffer unit 3, converts it into a parallel signal, and decodes this parallel signalized video signal in the variable length decoding unit.
It is sent to the inverse quantizer 5. The inverse quantization unit 5 restores the density of the quantization state of the input video signal to a uniform quantization state and outputs this to the inverse DCT unit. The inverse DCT unit 6 converts the input video signal into the time domain and then outputs it to the adder 22. The frame memory unit 7 stores the video signal of the previous frame, and outputs this to the adder 22 via the loop filter (LPF) unit 8 and the switch 23. The adder 22 receives the difference video signal from the inverse DCT unit 6 from the frame memory unit 7
It is added to the video signal of the previous frame read from to obtain a complete video signal that constitutes one screen, and this is output to the inverse format conversion unit 9. The inverse format converter 9 converts the input video signal into the NTSC system and outputs it to the D / A converter 10. The D / A converter 10 converts the input digital video signal into an analog signal and outputs it to a monitor or the like (not shown). Here, the switch 23 obtains information as to whether or not the video signal received from the variable length decoding unit 4 is a difference, and is turned on when the video signal is a difference, If you have not taken it, it will be turned off. Further, the variable length decoding unit 4 adjusts the strength of the edge portion of the received video signal by changing the characteristic of the loop filter unit 8 according to the control signal included in the received signal.

On the other hand, the voice signal is input to the voice delay unit 11, delayed by the count value by the counter unit 12, and then input to the PCM voice decoding unit 13 and the wideband voice decoding unit 14. The selection unit 15 selects an output signal from either the PCM audio decoding unit 13 or the wideband audio decoding unit 14 and outputs it to the D / A conversion unit 15 as an analog audio signal. It is output to the above-mentioned monitor section or the like. Here, the counter unit 12 is a down counter, and the audio signal is delayed by the count time of the counter unit 12. However, the down count value of the counter unit 12 is fixed and the delay amount of the audio signal is delayed. Is fixed.

By the way, the processing time of the video signal transmitted from the transmitting side differs depending on the quality of the image and the degree of change in the video. Therefore, when the video signal is input to the data separating section 1, the time shift amount thereof with respect to the audio signal is increased. Is not fixed and can change. Nevertheless, in the above-mentioned conventional apparatus, since the delay amount of the audio signal is fixed, the delay amount of the audio signal becomes inappropriate with respect to the video signal, and there is a drawback that the video and the audio do not match. It was

Further, in the conventional video conference apparatus, an image obtained from a television camera or the like may be subjected to image processing and transmitted to the receiving side. This image processing includes what is called a split function and a picture-in-picture function.

FIG. 9 is an explanatory view for performing the above-mentioned split function in the transmitting side device. Two TV cameras 91, 9
It is assumed that images of reference numerals 93 and 94 are taken in 2. When the upper 1/4 and lower 1/4 of the images 93 and 94 are cut off and the middle 1/2 is combined as shown by reference numeral 95 to form one image, which is transmitted to the receiving side, the receiving side Code 9
A composite image like 6 can be obtained.

FIG. 10 is an explanatory diagram for performing the picture-in-picture function in the receiving side device. The image received from the transmitting side device is copied as a parent screen 101, and a part of the child screen 102 is combined with the image of its own station and displayed on the monitor.

The above-described spirit function combines half images in the upper half and the lower half, and the picture-in-picture function combines the images of its own station at a predetermined position. The boundary between images is fixed, and there is a drawback in that the flexibility of image composition is lacking and the usability is poor for the user.

[0010]

In the conventional video conference apparatus, since the delay amount of the received audio signal is fixed, the video signal whose time shift amount changes depending on the image quality and the change of the image. However, there is a drawback that the output timing of the audio signal is not matched. In particular, when a subject or the like moves, it takes time to process the video signal on the transmission side, and audio is output first to the video, which gives a viewer a feeling of strangeness.

Further, in the conventional video conference apparatus, for example, two images taken by two cameras on the transmitting side are combined into one screen in the upper half and the lower half to be transmitted, or there is a predetermined received image. Although it is possible to combine the image of your own station at a certain position, the boundary between the image and the image is fixed in such image processing, and it is not always possible for the user to fit the image in the part that he wants to combine. There is a drawback that the image composition lacks flexibility.

Therefore, the present invention eliminates the above-mentioned drawbacks and changes the delay amount of the audio signal in accordance with the delay amount of the video signal so that the output timings of the video and audio are always matched to the image quality and the change of the image. It is an object of the present invention to provide a video conference apparatus capable of performing image composition by freely changing the boundary between images.

[0013]

According to a first aspect of the present invention, a television which receives a multiplexed video signal and audio signal on a transmission line, separates both signals, and performs a reception process on each separated signal. The conference apparatus has a configuration including a detection unit that detects the delay amount of the received video signal, and a delay unit that changes the delay time of the audio signal in accordance with the delay amount detected by the detection unit.

According to a fourth aspect of the present invention, there is provided a video conference apparatus having a function of performing image synthesis on a video signal and transmitting the image signal.
First specifying means for specifying an arbitrary part of the video signal for one screen, cutting means for cutting out the video signal part specified by the first specifying means, and among the non-composite video signals for one screen A second specifying means for specifying an arbitrary part of the above, and a combining means for combining the video signal cut out by the cutting means with the part specified by the second specifying means.

[0015]

According to the first aspect of the invention, the detecting means detects the delay amount of the received video signal. The delay means changes the delay time of the audio signal in accordance with the delay amount detected by the detection means. As a result, even if the video signal that is transmitted is delayed due to changes in image quality or image, the delay time of the received audio signal can be changed accordingly, so the output timing of the video and audio must always be matched. You can

In the video conference apparatus of the invention of claim 4,
The first designating means designates an arbitrary portion of the video signal for one screen. The cutout unit cuts out the video signal portion designated by the first designation unit. The second designating means designates an arbitrary part in the non-composite video signal for one screen. The synthesizing means synthesizes the video signal cut out by the cutting means with the portion designated by the second designating means. As a result, an arbitrary image can be cut and combined into an arbitrary portion, the degree of freedom in image combination is improved, and usability is significantly improved.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of the video conference apparatus of the first aspect of the invention. 1 is a data separation unit, 2
Is an error correction unit, 3 is an input buffer unit, 4 is a variable length decoding unit, 5 is an inverse quantization unit, 6 is an inverse DCT unit, 7 is a frame memory unit, 8 is a loop filter unit, and 9 is an inverse format conversion unit. Reference numeral 10 is a D / A conversion unit, 11 is a voice delay unit, 12 is a counter unit, 13 is a PCM voice decoding unit, 14 is a wideband voice decoding unit, 15 is a switching unit, and 16 is a D / A conversion unit. , 17 is a synchronizing signal generator, 18 is a chopping-off calculator, 1
Reference numeral 9 is an average value calculation unit, 20 is a delay amount calculation unit, and 21 is a comparison unit.

Next, the operation of this embodiment will be described. The data separation unit 1 inputs the video signal and the audio signal which are multiplexed and sent from the transmission line, separates these two signals, the video signal to the error correction unit 2 and the audio signal to the audio delay unit 11. Output. The error correction unit 2 corrects the error of the input video signal, sends it to the input buffer unit 3, decodes it in the variable length decoding unit, and then sends it to the dequantization unit 5. The inverse quantization unit 5 restores the density of the quantization state of the input video signal to a uniform quantization state and outputs this to the inverse DCT unit. Reverse D
The CT unit 6 converts the input video signal into the time domain and then outputs it to the adder 22. The frame memory unit 7 stores the video signal of the previous frame, and outputs this to the adder 22 via the loop filter (LPF) 8 and the switch 23. The adder 22 adds the differential video signal from the inverse DCT unit 6 to the video signal of the previous frame read from the frame memory unit 7 to form a complete video signal forming one screen, and then adds this to the format reverse conversion unit. Output to 9. The format reverse converter 9 converts the input video signal into the NTSC system and outputs it to the D / A converter 10. D
The / A converter 10 converts the input digital video signal into an analog signal and outputs it to a monitor (not shown) or the like. Here, the switch 23 obtains information as to whether or not the video signal received from the variable length decoding unit 4 is a difference, and is turned on when the video signal is a difference, If not taken off. Further, the variable length decoding unit 4 changes the characteristics of the loop filter unit 8 according to the control signal included in the received signal,
The strength of the edge of the received video signal is adjusted.

The sync signal generator 17 generates a TR sync signal 50 and an arithmetic sync signal 60 from the picture header section separated by the variable length decoder 4, and drops the sync signals into a chopping operation section 18, It is output to the average value calculator 19. The drop-off calculator 18 receives the TR signal from the variable-length decoder 4 according to the synchronization signal 50, calculates the difference between the previous frame and the frame for each frame, and calculates the number of drops of the received video signal. The obtained number of dropped frames is output to the average value calculation unit 19. The average value calculation unit 19 adds the number of dropouts input by the synchronization signal 60 n times, divides the addition result by n, obtains the average value of the number of dropouts, and outputs this to the delay amount calculation unit 20. . The delay amount calculation unit 20 obtains the delay amount of the audio signal from the average value of the input dropping numbers, and sets it in the counter unit 12.

On the other hand, the voice signal is input to the voice delay unit 11 and delayed by the count value of the counter unit 12, and then P
It is input to the CM voice decoding unit 13 and the wideband voice decoding unit 14. The switching unit 15 selects an output signal from either the PCM audio decoding unit 13 or the wideband audio decoding unit 14 and outputs it to the D / A conversion unit 16, where it is converted into an analog audio signal as described above. Output to the monitor unit etc. Here, the counter unit 12 is a down counter from the set value by the delay amount calculation unit 20, and the audio signal is delayed by the count time of the counter unit 12. The analog audio signal output from the D / A conversion unit 16 is input to the comparison unit 21, the level thereof is compared with the reference value 0, and when the level becomes equal to or less than the reference value for a certain period of time, it is output to the delay amount calculation unit 20. Then, the delay amount output from the delay amount calculation unit 20 is set to a fixed value D, and this is set to the counter unit 1
Set to 2.

FIG. 2 is a block diagram calculator 18 shown in FIG.
3 is a circuit diagram showing a detailed example of an average value calculator 19 and a delay amount calculator 20. FIG. The synchronization signal generator 17 synchronizes the internal clock as shown in FIG. 3A with the frame pulse as shown in FIG. 3B input from the variable length decoder 4.
Synchronous signals 50 and 60 as shown in FIGS. 3C and 3D are created from this internal clock, and these are dropped and output to the latch circuit 181 of the arithmetic unit 18. Latch circuit 18
When the synchronization signal 50 is input, 1 latches the TR input from the variable length decoding unit 4 and outputs it to the subtractor 182. Therefore, the subtractor 182 receives the T that is input this time.
T of the previous frame input from the latch circuit 181 from R
R is subtracted and the obtained difference is output to the adder 191 of the average value calculation unit 19 (corresponding to the amount of frame drop). When the synchronization signal 60 is input, the latch circuit 191 of the average value calculation unit 19 latches the added value output from the adder 192,
This is output to the adder 192.

As a result, the adder 192 skips this and sequentially adds the differences input from the calculation unit 18, but the latch circuit 191 resets when the reset signal 70 is input from the synchronization signal generation unit 17. Since the latch circuit 191 is reset, the adder 192 adds the difference input between the two. Sync signal generator 1
7 generates a reset signal 70 as shown in FIG. 3 (F) every time the frame pulse is input n times as shown in FIG. 3 (E), and outputs it to the latch circuit 191.
Therefore, the adder 192 adds TR n times,
The added value of the number of times is input to the divider 201 of the delay amount calculation unit 20. The division unit 201 of the delay amount calculation unit 20 divides the added value of n times by n and outputs an average value of the difference n times, inputs this to the multiplication unit 202, and multiplies the average value by the coefficient A. Then, it is added to the fixed value D separately input to this and output to the counter unit 12 of FIG. When the multiplication unit 202 receives a signal from the comparison unit 21 indicating that the audio signal level has become 0 for a certain period of time or more, it sets the delay amount to the fixed value D and outputs it to the counter unit 12. It should be noted that there is a correspondence between the amount of drop of the top and the amount of information of the transmitted video signal. A large amount of drop of the top means that the amount of information of the video signal is large, and the processing time of the video signal when transmitting is long. Corresponding to this, a small amount of top drop corresponds to a small amount of information of the video signal and a short processing time of the video signal at the transmission side. Since the video signal is delayed by that much, if the delay amount of the audio signal is increased and the amount of top drop is small,
Since the video signal is not delayed by that much, if the delay amount of the audio signal is reduced, the timing of the video and the audio can be matched.

FIG. 4 is a characteristic diagram of the delay amount calculated by the delay amount calculating section 20 described above. The horizontal axis represents the average number of dropped frames, and the vertical axis represents the audio delay amount. , If the average value of dropped frames is less than or equal to h, the delay amount calculation unit 20 determines that y = Ax + D.
The delay amount is calculated according to (A = 33 ms), and the characteristic becomes a. However, when the average value of dropped frames becomes equal to or greater than h, the delay amount calculation unit 20 calculates the delay amount according to y = x + D, resulting in the characteristic of b. When the comparison unit 21 determines that there is no audio signal for a certain period of time or more, the delay amount calculation unit 20 sets the delay amount to the fixed value D, resulting in the characteristic of C.

According to this embodiment, the frame drop amount is calculated from the received video signal, the delay amount of the audio signal is calculated from the average value of the frame drop amount over n frames, and the delay amount is calculated by the counter unit 12. By changing the delay time of the audio signal by the audio delay unit 11 in accordance with the amount of frame drop, that is, when the amount of frame drop is large and the video signal is delayed, the delay of the audio signal is delayed. If the video signal is not delayed by increasing the amount of time drop and conversely, by shortening the delay time of the audio signal, even if there is a large change in image quality or image, it is possible to match the delay of the video signal. The delay amount of the audio signal can also be changed, and the video and audio can always be matched.

FIG. 5 is a block diagram showing another embodiment of the video conference apparatus of the invention of claim 4. Reference numeral 31 is a television camera, 32 is an input frame memory, 33 is an address counter control section which constitutes an image cut control system, 34 is a cut image position designation section which constitutes the control system, and 35-1 to 35-3 are enlarged images. Enlargement / reduction unit for reducing, 36 is a background memory for storing image background data, 37 is an output frame memory,
38 is an address counter control unit, 39 is a composite position designation unit, 40 is an image compression / expansion device, 41 is a monitor, and 42 is a transmission path. Among these, between the input frame memory 32 and the output frame memory 37, the cutout of the image, the enlargement / reduction of the cutout image, and the image combination are performed.

Next, the operation of this embodiment will be described. One frame of image data showing the state of the conference captured by the television camera 31 is stored in the input frame memory 32.
Since the image data in the input frame memory 32 is displayed on a monitor (not shown), when the user designates the image portion to be sent on the screen of the monitor with a keyboard or a mouse, this image designation information designates the cut image position. It is input to the unit 34. The cut-out image position designation unit 34 determines the start load value of the horizontal and vertical counters that generate the read address according to the cut-out designation information, and controls the address counter control unit 33 based on this to determine the load value from the input frame memory 32. The designated image portion is read and output to the scaling units 35-1 to 35-3. When there are a plurality of image parts to be cut out, the cut-out image position specifying part 34 controls the address counter control part 33 by the number of cut-outs to control the read address of the input frame memory.

The enlarging / reducing units 35-1 to 35-3 perform enlarging / reducing of the image portion cut out from the input frame memory 32. This enlarging / reducing is performed in the vertical direction of the image input by line conversion or pixel conversion. It is performed in the horizontal direction. FIG. 6 is a block diagram of increasing / decreasing the number of lines for performing the enlargement / reduction processing in the vertical direction. The image data cut out from the input frame memory 32 is added with the coefficient calculated by the coefficient calculation unit 63 as it is, and is temporarily stored in the 1-line memory 61 together with the data input to the adder 66.
Data that is read one line after one, added with a coefficient by the coefficient calculation unit 64 and input to the adder 66, and one line of image data output from the one line memory 61 are temporarily stored in the one line memory 62. Okay, this memory 6
There is data to be input to the adder 66 after being read out after one line from 2 and added with the coefficient by the coefficient calculation unit 65. As a result, the adder 66 outputs data without delay, 1
By calculating the line delay data and the two line delay data, the number of lines of the image data in the vertical direction is increased or decreased to enlarge or reduce the image in the vertical direction. The enlarged or reduced image data thus obtained is written in a designated position in the output frame memory 37. Vertical scaling is performed by using two line memories as shown in FIG. 6, but horizontal scaling is performed by using two D flip-flops so that data without delay and data with one pixel delay are provided. This is performed by creating data with a delay of 2 pixels and calculating the data to increase or decrease the number of pixels.

By the way, the background image data stored in the background memory 36 is written in the output frame memory 37 in advance and the user is input from the enlargement / reduction unit 35 by the time the above-mentioned enlargement / reduction processing is completed. The position where the image data is to be fitted is designated in the composite position designation unit 39 with a mouse or the like. In this case as well, an image for one screen in the output frame memory 37 is displayed on a monitor (not shown), and the position to be fitted is specified on this monitor. As a result, the composite position designation unit 39 determines the start load values of the horizontal and vertical counters for the write address, and controls the address counter control unit 38 to write the image data at the position designated by the user. To output to the output frame memory 37. As a result, the image data output from the enlarging / reducing unit 35 is output to the output frame memory 3 by the address.
The data is written in a designated position in 7 and image synthesis is performed. The address counter control unit 38 generates write addresses for only the images to be combined, and sequentially writes the corresponding image data sent from the enlarging / reducing unit 35 to each position designated by the user.

When the writing of the image data into the output frame memory 37 as described above is completed, the image data is read from the output frame memory 37 by vertical scanning, and the read image data is sent to the image compression / expansion device 40. . At the same time, the background data in the background memory 36 is overwritten in the output frame memory 37 to prepare for the next frame.

FIG. 7 is a diagram showing a process of image composition in the output frame memory 37 described above, which will be described with reference to a screen diagram when image data in the output frame memory 37 is displayed on a monitor screen. When the image data in the background memory 36 is written in the output frame memory 37, the screen is only the background as in step 701. When the cut-out image 1 is overwritten in the output frame memory 37 which stores this background image data, step 702 is performed. Further, when the cutout image 2 is overwritten on the output frame memory 37, step 703 is performed,
When the cutout image 3 is overwritten in the same manner, step 704 is performed. After that, output frame memory 3
When the composite image data is read from 7, the screen data as in step 705 is read, and the image data is compressed by the image compression / expansion device 40 and sent to the transmission path 42. The background image data from the background memory 36 is written in the output frame memory 37 again, and the process returns to step 701.

According to the present embodiment, the image data to be combined and transmitted can be arbitrarily designated, and the designated image data portion can be enlarged or reduced to be combined with an arbitrary portion of the transmitted image and transmitted. Since it is possible to freely set the boundary between the image to be combined and the combined image, the user can freely insert the above-mentioned arbitrary image into the portion to be inserted, and the degree of freedom of image combination is increased. It can be significantly improved. With this, it is possible to inform the other station of the content that the user really wants to transmit. Further, the importance of the image can be expressed by scaling the transmission image.
In addition, although information may be newly generated due to flicker of a fluorescent lamp or the like at the camera input, in this example, the background memory 36 is used.
Since there is used, there is an advantage that new information is not generated due to flicker, and that amount can be allocated to the increase in the number of frames.

[0032]

As described above, according to the video conference apparatus of the present invention, by changing the delay amount of the audio signal in accordance with the delay amount of the video signal, the video and the audio are constantly changed with respect to the image quality and the change of the image. The output timing can be matched, and the image composition can be performed by freely changing the boundary between the images.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a video conference device of the present invention.

FIG. 2 is a block diagram showing a detailed example of a top drop calculating unit, an average value calculating unit, and a delay unit calculating unit shown in FIG.

FIG. 3 is a time chart explaining the operation of FIG.

FIG. 4 is a calculation characteristic diagram of the delay amount calculation unit shown in FIGS. 1 and 2.

FIG. 5 is a block diagram showing another embodiment of the video conference apparatus of the present invention.

6 is a block diagram showing a configuration example of increasing / decreasing the number of lines in the vertical direction of the enlargement / reduction unit shown in FIG.

FIG. 7 is a diagram illustrating an example of a process of image synthesis in the apparatus of FIG.

FIG. 8 is a block diagram showing an example of a conventional video conference apparatus.

FIG. 9 is an explanatory diagram illustrating a conventional image synthesizing method.

FIG. 10 is a diagram showing an example of an image combined by a conventional image combining method.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Data separation part 2 ... Error correction part 3 ... Input buffer part 4 ... Variable length decoding part 5 ... Inverse quantization part 6 ... Inverse DCT part 7 ... Frame memory part 8 ... Loop filter part 9 ... Inverse format conversion part 10 , 16 ... D
/ A conversion unit 11 ... Voice delay unit 12 ... Counter unit 13 ... PCM voice decoding unit 14 ... Wideband voice decoding unit 15 ... Switching unit 17 ... Synchronization signal generating unit 18 ... Cover dropping calculation unit 19 ... Average value calculation Part 20 ... Delay amount calculation part 21 ... Comparison part 31 ... Television camera 32 ... Input frame memory 33, 38 ... Address counter control part 34 ... Cut image position designation part 35-1 to 35-3 ... Enlargement / reduction part 36 ... Background memory 37 ... Output frame memory 39 ... Compositing position designation unit 40 ... Image compression / expansion device 41 ... Monitor 42 ... Transmission path

Claims (6)

[Claims] 1. A video conferencing apparatus that receives a multiplexed video signal and audio signal on a transmission line, separates the two signals, and then performs a reception process on each separated signal. The video conferencing apparatus is provided with a detecting means for detecting the delay time, and a delay means for changing the delay time of the audio signal in accordance with the delay amount detected by the detecting means. 2. The video conferencing apparatus according to claim 1, wherein the detecting means calculates a dropout amount of the received video signal, and calculates the delay amount based on the dropout amount. 3. The video conference apparatus according to claim 1, wherein the delay means sets the delay amount to a fixed value when the level of the received audio signal is equal to or more than a predetermined time and equal to or less than a predetermined value. 4. A video conferencing apparatus having a function of performing image synthesis on a video signal and transmitting the first video signal, a first designating means for designating an arbitrary part of the video signal for one screen, and
Cutting-out means for cutting out the video signal portion designated by the designating means, second designation means for designating an arbitrary portion of the non-composite video signal for one screen, and the video cut-out by the clipping means. A video conference apparatus comprising: a synthesizing unit that synthesizes a signal into a portion designated by the second designating unit. 5. The video conference apparatus according to claim 4, further comprising enlarging / reducing means for enlarging or reducing the video signal portion cut out by the cutting means. 6. The video conference apparatus according to claim 4, further comprising a storage unit that holds the combined video signal for one screen.