JP2006129401A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce image output delay and to enhance amusement effects, in photographing or editing moving images. <P>SOLUTION: The image processing apparatus comprises one memory B, to which an input image is to be inputted by an image input means 100; a memory A, on the prestage of the one memory B, for storing the information of processing to be performed by a region detecting means 101, a graphics computing means 102 and an image compositing means 103; and the memory A on the prestage that completes processing or a memory, further on the prestage of the memory A of the prestage, for storing an image composited by the image compositing means 103. There is provided a motion detection means 106 for detecting the motion of a target that corrects the position determined by the graphics computing means 102, using motion information. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a moving image recording device such as a digital video camera or a mobile phone with a camera capable of shooting moving images, and an image processing device in a recorded moving image editing device such as a VTR or a DVD recorder.

  When shooting video with a digital video camera or mobile phone, shooting with a videophone, and editing with a VTR, DVD recorder, etc., not only shooting and editing, but also shooting or playback images A system that enhances the amusement effect has been considered by performing shooting and editing while synthesizing graphics (figures and images) such as a hat and a bag following the movement of a person or an object.

  Patent Document 1 shows an example of this system. In this system, as shown in FIG. 10, an image input unit 200, a face region detection unit 201 that detects a face region from an image, and a face component position are displayed. A face part detection unit 202 to detect, a graphic calculation unit 203 for determining a position for superimposing graphics on the input image, a graphic synthesis unit 204 for superimposing the input image and graphics, and the synthesized image The image output unit 205 is configured to output. With this configuration, a face area is detected from a captured image, and graphics are synthesized following the face area, thereby improving image effects such as amusement and presentation.

  In such a case, in general, the processing in the face area detection unit 201 and the face part detection unit 202 for detecting a specific area and part such as a face, mouth, and nose from an image requires extremely high-speed arithmetic processing. To do. This is because when the calculation processing speed for detecting the face area and the parts is slow, the image capture of the next frame from the image input unit 200 may be completed during the detection. Therefore, in general, instead of performing the high-speed arithmetic processing described above, the buffer memory for storing the input image is increased to temporarily store the input image, and the arithmetic processing speed of the face area detection unit 201 or the like is reduced. A method of covering the thickness is taken.

  In this way, it is possible to increase the buffer memory so that the processing in the image input unit 200, the face area detection unit 201, the figure calculation unit 203, the figure synthesis unit 204, and the image output unit 205 can be performed smoothly. Assuming the configuration of FIG. 10 as described above, FIG. 11 shows that the processing by the area detecting means corresponding to the face area detecting section 201, the figure calculating means corresponding to the figure calculating section 203, and the image combining means corresponding to the figure combining section 204 is completed smoothly and quickly. The case where it does is illustrated. In FIG. 11, three periods are shown as a time axis, and in each period, data is exchanged with respect to the three buffer memories A, B, and C by the image input means, the area detection means, the figure calculation means, the image composition means, and the image output means. It shows a state. In FIG. 11, the processing time of each means (each cycle in FIG. 11) is the processing time required for the image input means: the total processing time required for the area detection means, the figure calculation means, and the image composition means: the image output processing means. The required processing time ratio is 1: 1: 1.

  That is, first, in the first period, the image input means inputs an image into the memory A. In the second period, the image input means inputs an image to the memory B. At the same time, the area detecting means performs area detection using the image input to the memory A, the figure calculating means calculates the combining position, and the image combining means combines the images in the memory A. In the third period, the image input means inputs an image to the memory C. At the same time, the area detecting means performs area detection using the image input to the memory B, the figure calculating means calculates the combining position, and the image combining means combines the images in the memory B. Further, the image output means outputs a composite image in the memory A. By such an operation of three cycles, smooth processing can be realized by the image buffer memory of the three surfaces. For example, the image input input to the memory A is image-synthesized and output in the third cycle. It will be.

  Next, in FIG. 12, the processing of the area detection means is heavy, and the total processing time required for the area detection means, the graphic calculation means, and the image synthesis means exceeds the processing time required for the image input means or image output means. Is illustrated. For this reason, in this example, the area detection means, the figure calculation means, and the image composition means are assigned to different memories. That is, in this example, the processing time of each means is the processing time required for the image input means: the total processing time required for the area detection means and the graphic calculation means: the processing time required for the image composition means: the processing time required for the image output processing means. The ratio is 1: 1: 1: 1.

First, in the first cycle, the image input means inputs an image to the memory A. In the second period, the image input means inputs an image to the memory B. At the same time, the area detection means performs area detection using the image input to the memory A, and the figure calculation means calculates the composite position. In the third period, the image input means inputs an image to the memory C. At the same time, the area detecting means performs area detection using the image input to the memory B, and the figure calculating means calculates the composite position. Further, the image synthesis means synthesizes the image in the memory A. In the fourth period, the image input means inputs an image to the memory D. At the same time, the area detection means performs area detection using the image input to the memory C, and the figure calculation means calculates the composite position. Further, the image synthesizing unit synthesizes the image in the memory B. Then, the image output means outputs the composite image in the memory A. As described above, by using the four-cycle image buffer memory, it is possible to smoothly realize the image processing even if the area detecting unit becomes heavy by using the four-screen image buffer memory.
JP 2000-322588 A

  However, the conventional image processing apparatus has the following problems. That is, in FIG. 11, three cycles are required until the graphics are combined with the input image from the image input means and output by the image output means, whereas the processing of the area detection means becomes heavy. 12 requires four cycles. This means that the period from image input to image output requires three cycles. This delay in image output can be as long as 3 cycles. For example, if digital video is being shot, it is between the image that is directly viewed by the person being photographed and the image that is being viewed from the viewfinder or LCD screen. This delay can no longer be ignored, creating a sense of incongruity. In addition, in a video phone or the like, a so-called lip sync shift occurs in which the image and the sound cannot be synchronized with the image delay, and for example, the operation of the mouth of the sound and the image does not match.

  The present invention has been made to solve such a problem. When a photograph or reproduction image is followed by a person or an object (target object) and a graphic (compositing target) is combined to perform shooting or editing. Another object of the present invention is to provide an image processing apparatus with little delay in image output when image synthesis is performed on an input image.

  The image processing apparatus of the present invention determines an object to be synthesized and a position in accordance with an image input means, an area detection means for detecting an object from an input image by the image input means, and an object detected by the area detection means. And an image compositing means for compositing an image based on the position determined by the graphic calculating means when compositing the composition target determined by the graphic calculating means for the input image by the image input means, In the image processing apparatus having an image output means for outputting a composite image by the image composition means, and having a memory arranged in order as a buffer for each means, an input image is inputted by the image input means. A memory for storing processing information performed by the memory, the area detecting means, the graphic calculating means, and the image synthesizing means; Taking a pre-stage of the memory, and stores the composite image by the image synthesizing means, the processing and further preceding memory than the previous stage of the memory or the front of the memory that has been completed, the configuration provided with the.

  According to the present invention, there is provided an image processing apparatus that has a delay in image output and an enhanced amusement effect because the period from image input to image output is shortened compared to the prior art. It has an excellent effect in that it can be performed.

  In addition, by correcting the composite coordinates based on the motion detection information, an image processing apparatus is provided that has less image output delay than the conventional technology, can follow the composite image with time, and has improved the amusement effect. It has an excellent effect in that it can be used.

  Furthermore, by controlling whether or not to correct the composite coordinates, it is possible to provide an image processing apparatus that has less image output delay, can control the follow-up correction of the composite image, and further enhances the amusement effect. In this respect, it has an excellent effect.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 shows an image processing apparatus according to Embodiment 1 of the present invention. In FIG. 1, an image input unit 100 stores image data in a memory 105 (any one of memories A, B, C, and D). The image data may be image data from an input device such as a camera, image data decoded by a decoding device such as a TV tuner, a DVD player, or a VTR, or downloaded from a network. It may be image data. Further, it may be image data that has undergone color conversion (YUV-RGB conversion, etc.), color increase / decrease processing, D / A conversion processing, or the like.

  The area detection unit 101 detects an area of an object to be detected from the input image stored by the image input unit 100. The detection target object (target object) may be the whole human being, a human face, or a partial region such as a hand, foot, eye, nose, ear, or mouth. Further, not only humans but also animals, and artificial objects such as cars and trains may be used. As a method for detecting these objects, there is a method in which a template generated from an object collected in advance is applied to each part of an image and a part having a high correlation with the template is detected as an object region. In addition, the object may be detected using a distance image from another input device, a range finder (distance alignment finder) capable of measuring distance, and the object is detected using information from the temperature sensor. May be.

  The graphic calculation unit 102 determines graphics (target object) to be combined and composite coordinates (position) for the region detected by the region detection unit 101. For example, if it is a nose, it will synthesize a frog graphic, if it is an eye, select a graphic such as glasses, and if it is a head, select graphics such as a hat. Tell.

  The image composition unit 103 synthesizes the image input by the image input unit 100 and the graphics selected by the graphic calculation unit 102 based on the composite coordinates determined by the graphic calculation unit 102 and stores the image in the memory 105. . Note that color conversion (YUV-RGB conversion, RGB-YUV conversion, etc.), color increase / reduction processing, or the like may be performed when storing in the memory 105.

  The image output means 104 receives the composite image stored by the image composition means 103 and outputs it to a display device such as an LCD. Note that the image output method may be converted to the NTSC or PAL format to enable output to the TV screen.

  Next, a specific operation will be described with reference to FIG. 2 according to the configuration diagram of FIG. FIG. 2 has the same configuration as FIG. 12 described above. In this example, the time axis is shown from 1 to 5 cycles, and the processing time of each means is the same as in FIG. The ratio of (processing time required) :( total processing time required for the area detection means 101 and graphic calculation means 102) :( processing time required for the image composition means 103) :( processing time required for the image output means 104) is 1: 1: 1: 1.

  First, in the first period, the image input means 100 inputs an image to the memory A. In the second period, the image input means 100 inputs an image to the memory B. At the same time, the area detection unit 101 performs area detection using the image input to the memory A, and the figure calculation unit 102 calculates the composite coordinates (position). In the third period, the image input unit 100 inputs an image to the memory C. At the same time, the area detection means 101 performs area detection using the image input to the memory B, and the figure calculation means 102 calculates the position. Furthermore, the image composition means 103 does not synthesize the image of the memory A and the graphic, but synthesizes the graphic with the image of the memory B. The composite image is stored in the memory A. Here, the memory A stores not the image input in the first cycle but the result of the synthesis process performed on the image input in the second cycle. This is because the area detection unit 101 has not completed the area detection process on the image in the memory B, and thus the image in the memory B cannot be overwritten until the process is completed. If processing is completed, control may be performed to overwrite the memory B.

  In the fourth period, the image input unit 100 inputs an image to the memory D. At the same time, the area detecting means 101 performs area detection using the image input to the memory C, and the figure calculating means 102 calculates the position. Further, the image synthesis unit 103 synthesizes graphics with the image in the memory C and stores the synthesized image in the memory B. Then, the image output means 104 outputs the composite image in the memory A. Similarly, also in the fifth cycle, the image input means 100 inputs an image to the memory A, and at the same time, the area detection means 101 performs area detection using the image input to the memory D, and the figure calculation means 102 determines the position. Further, the image composition means 103 synthesizes graphics with the image in the memory D and stores the composite image in the memory C. Then, the image output means 104 outputs the composite image in the memory B.

  As described above, the image processing apparatus according to Embodiment 1 of the present invention can output a synthesized image obtained by synthesizing an image input in the second period in the fourth period, as shown in FIG. In the conventional example shown, it takes three cycles from image input to image output, but in this embodiment, it can be shortened to two cycles. Therefore, it has an excellent effect in that it can provide an image processing device that has less delay in image output than the conventional technology and has enhanced amusement effect.

  In the first embodiment of the present invention, the example in which the latest input image from the image input unit 100 is synthesized is described in mind. However, the old and new input images are selected and extracted and synthesized. It is also possible to do so.

  The image processing apparatus according to the first embodiment can be incorporated into, for example, a portable terminal, and can be realized as a mobile communication system including the image processing apparatus. The image processing apparatus according to the first embodiment can be incorporated into, for example, a portable terminal, and can be realized as an image editing system including the image processing apparatus.

(Embodiment 2)
FIG. 3 shows an image processing apparatus according to the second embodiment of the present invention. In FIG. 3, the same components as those in the first embodiment shown in FIG.

  The motion detection means 106 connected to the figure calculation means 102 detects the motion of the detected object based on the object detected by the area detection means 101. The movement of the object can be detected by using synthetic coordinates calculated by the graphic calculation means 102 for each period. For example, as shown in FIG. 4A, the composite coordinates determined by the graphic calculation means 102 in the second period are (3, 3), and the composite coordinates determined by the graphic calculation means 102 in the third period are shown in FIG. In the case of (5, 2) as in 4 (b), the movement is a vector (2, -1) in the component display vector. The motion detection information detected in this way is transmitted to the image composition means 103 connected to the motion detection means 106.

  The image synthesizing unit 103 synthesizes the coordinates (position) determined by the graphic calculation unit 102 when combining the graphic or image to be combined determined by the graphic calculation unit 102 with the latest input image input by the image input unit 100. ) And the motion information detected by the motion detection means 106, the composite position is corrected and combined. In the fourth period of FIG. 2, the image composition unit 103 uses the vector (2, -1) detected by the motion detection unit 106 based on the composite coordinates (5, 2) determined by the figure calculation unit 102. Correction is performed, and the composite coordinates are set to (7, 1) as shown in FIG. In this way, the input images are often temporally correlated, and the input images input in the third period stored in the memory C are input in the first and second periods. This is because the motion information detected between the input images tends to be similar to or infinitely close to the motion information detected between the input images input in the second and third periods.

  By synthesizing the image with the composite coordinates (7, 1) corrected in this way, as shown in FIG. 5A, for example, the bow tie does not correctly follow the neck in the case of no correction. In 5 (b), it follows correctly.

  As described above, the image processing apparatus according to the second embodiment of the present invention has an image output delay that is smaller in image output delay than the conventional technique, can follow a composite image with time, and has an enhanced amusement effect. It has an excellent effect in that it can be provided.

  In the second embodiment, the example in which the latest input image from the image input means 100 is synthesized is described in mind. However, the new and old input images are selected and extracted and synthesized. It is also possible to do.

  In the second embodiment, as the motion detection method, the motion information is detected using only the previous input image and the previous input image as shown in FIG. 4, but the motion information is also detected using the previous input information. Detection can also be performed.

  In addition, the motion information in the second embodiment is calculated based on the information from the graphic calculation means 102, but is not limited to this if the motion information is obtained. For example, motion information obtained from an encoding device or a decoding device may be used in a device obtained by adding an encoding device such as MPEG-4 and / or a decoding device to the image processing device of the present invention.

  The image processing apparatus according to the second embodiment can be incorporated into, for example, a portable terminal, and can be realized as a mobile communication system including the image processing apparatus. Further, the image processing apparatus according to the second embodiment can be incorporated into, for example, a portable terminal, and can be realized as an image editing system including the image processing apparatus.

(Embodiment 3)
FIG. 6 shows an image processing apparatus according to the third embodiment of the present invention. In FIG. 6, the same reference numerals are given to the same components as those in the first embodiment shown in FIG. 1 or the second embodiment shown in FIG. 3, and the description thereof is omitted.

  The composition correction control means 107 connected to the image composition means 103 controls the image composition means 103 to correct the composition position. The synthesis of the input image from the image input means 100 and the graphics should completely follow the object as illustrated in FIG. 5, but it is reversed by consciously delaying the tracking to some extent. In some cases, the amusement effect can be enhanced. For this reason, the synthesis correction control means 107 is provided so that the correction control can be performed in the synthesis with the graphics. As a specific example, for example, in cartoons and animations, it is interesting to shift the wig and shift the wig. Here, the correction of the composite position may be, for example, control that does not take into account motion information by the motion detection unit 106 and control that adds or subtracts the degree of addition of motion information.

  As described above, the image processing apparatus according to the third embodiment of the present invention provides an image processing apparatus that has a small image output delay, can control the follow-up correction of the composite image, and further enhances the amusement effect. It has an excellent effect in that it can.

  In the third embodiment, the example in which the latest input image from the image input unit 100 is synthesized has been described in mind. However, the new and old input images are selected and extracted and synthesized. It is also possible to do.

  In addition, the image processing apparatus according to the third embodiment can be incorporated in, for example, a portable terminal, and can be realized as a mobile communication system including the image processing apparatus. The image processing apparatus according to the third embodiment can be incorporated into, for example, a portable terminal, and can be realized as an image editing system including the image processing apparatus.

(Embodiment 4)
FIG. 7 shows a flowchart of Embodiment 4 which is an image processing method of the present invention. In FIG. 7, first, an image is input by image input processing (ST701). Next, an area detection process (ST702) is performed, and based on the detection result, a graphic calculation process (ST703) is performed to calculate graphics and a synthesized coordinate of the graphic. In the image composition process (ST704), graphics are synthesized with the latest input image based on the calculated coordinates, and the image is output by the image output process (ST705). When each process is completed, the process is repeated. Each process may be synchronized or may be asynchronous. In each process, a memory to be used is a buffer memory juxtaposed in a plurality of orders as shown in FIG. Here, when the image input process of ST701 is performed on a memory in a certain cycle, the area detection process of ST702, the graphic calculation process of ST703, and the image composition process of ST704 are executed for the memory in the next period, The image is recorded in the certain memory that has been processed or the preceding memory of the certain memory, and then the stored image in the certain memory or the preceding memory is output in the image output process of ST705 in the next cycle.

  As described above, in the image processing method according to the fourth embodiment of the present invention, in the example of FIG. 2, in the fourth period, a composite image obtained by performing the composite process on the image input in the second period is output. In the prior art shown in FIG. 12, the period from the input of an image to the output of the image required three cycles is shortened to two. Therefore, it has an excellent effect in that it can provide an image processing method in which an image output delay is small as compared with the conventional technique and an amusement effect is enhanced.

  In the fourth embodiment, the example of combining the latest input image in the image input process has been described in mind. However, the new and old input images are selected and extracted and combined. It is also possible.

  In the fourth embodiment, the image processing method can also be realized by software that can be executed by a computer. The software is stored in a computer-readable recording medium and read from the recording medium. Can also be realized.

(Embodiment 5)
FIG. 8 is a flowchart of the image processing method according to the fifth embodiment of the present invention. In FIG. 8, the processes with the same reference numerals as those shown in FIG. 7 are the same processes and will not be described. In FIG. 8, after the figure calculation process (ST703), a motion detection process (ST801) is performed. In this motion detection process (ST801), the motion of the object is detected based on the composite coordinates. In the image composition process (ST704), the composition coordinates are corrected based on the motion detection information, the image is composed, and the image is output by the image output process. When each process is completed, the process is repeated. Each process may be synchronized or may be asynchronous. In each process, a memory to be used is a buffer memory juxtaposed in a plurality of orders as shown in FIG. Also in this case, when the image input process of ST701 is performed on a memory in a certain cycle, the area detection process of ST702, the graphic calculation process of ST703, the motion detection process of ST801, and the image of ST704 are performed for the memory in the next cycle. The composite processing is executed, and the composite image is recorded in the certain memory that has been processed or the previous memory of the certain memory, and then the stored image in the certain memory or the previous memory in the next cycle is subjected to the image output processing of ST705. Output.

  As described above, the image processing method according to the fifth embodiment of the present invention is an image processing apparatus that has a smaller image output delay than the prior art, can follow a composite image with time, and has improved an amusement effect. It has an excellent effect in that it can be provided.

  In the fifth embodiment, the image processing method can also be realized by software that can be executed by a computer. The software is stored in a computer-readable recording medium and read from the recording medium. Can also be realized.

(Embodiment 6)
FIG. 9 shows a flowchart of an image processing method according to the sixth embodiment of the present invention. In FIG. 9, the processes with the same reference numerals as those shown in FIGS. 7 and 8 are the same processes and will not be described. In FIG. 9, after the motion detection process (ST801), in the image composition process (ST704), whether or not to correct the composite coordinates is controlled in the composite correction control process (ST901). Based on the control information, the composite coordinates are corrected and the image is composited. Thereafter, an image is output by image output processing (ST705). When each process is completed, the process is repeated. Each process may be synchronized or may be asynchronous. In each process, a memory to be used is a buffer memory juxtaposed in a plurality of orders as shown in FIG. Also in this case, when the image input process of ST701 is performed on a memory in a certain cycle, the area detection process of ST702, the graphic calculation process of ST703, the motion detection process of ST801, and the synthesis of ST901 are performed for the memory in the next cycle. The image synthesis process of ST704 accompanied by the correction control process is executed, and the synthesized image is recorded in the memory that has completed the process or the previous memory of the memory, and then stored in the certain memory or the previous memory in the next cycle. The image is output by the image output process of ST705. Here, the correction of the composite position can be, for example, control that does not take motion information by motion detection processing (ST801) at all, and control that adds or subtracts the degree of addition of motion information.

  As described above, the image processing method according to the sixth embodiment of the present invention has less image output delay than the conventional technique, enables control of the follow-up correction of the composite image, and further enhances the amusement effect. This has an excellent effect in that an image processing apparatus can be provided.

  In the sixth embodiment, the image processing method can also be realized by software that can be executed by a computer. The software is stored in a computer-readable recording medium and read from the recording medium. Can also be realized.

  The present invention is suitable for use in moving image shooting or editing devices that increase the amusement effect.

1 is a block diagram showing a configuration of an image processing apparatus according to Embodiment 1 of the present invention. Explanatory drawing showing the operation timing in the first embodiment Block diagram showing a configuration of an image processing apparatus according to Embodiment 2 of the present invention. Explanatory drawing of the movement of the composite coordinates in the second embodiment Explanatory drawing of the synthesis result in Embodiment 2 Block diagram showing a configuration of an image processing apparatus according to Embodiment 3 of the present invention. FIG. 9 is a flowchart showing processing of an image processing method according to Embodiment 4 of the present invention. FIG. 7 is a flowchart showing processing of an image processing method according to Embodiment 5 of the present invention Flow chart showing processing of an image processing method in Embodiment 6 of the present invention Block diagram showing a conventional image processing apparatus Explanatory drawing showing operation timing of conventional technology Explanatory drawing showing operation timing of conventional technology

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Image input means 101 Area | region detection means 102 Graphic calculation means 103 Image composition means 104 Image output means 105 Memory 106 Motion detection means 107 Composition correction control means

Claims (9)

Image input means, area detection means for detecting an object from an input image by the image input means, figure calculation means for determining a composition target and position according to the object detected by the area detection means, and the image When synthesizing the synthesis target determined by the graphic calculation means with respect to the input image by the input means, the image synthesis means for synthesizing the image based on the position determined by the graphic calculation means, and the composite image by the image synthesis means In an image processing apparatus having an image output means for outputting, and having a memory juxtaposed in order as a buffer for each means,
A memory in which an input image is input by the image input unit; a memory preceding the one memory for storing processing information performed by the region detection unit, the graphic calculation unit, and the image synthesis unit; An image processing apparatus comprising: the previous-stage memory that has completed the processing, or a previous-stage memory that stores the composite image by the image synthesis means;   When the synthesis target determined by the graphic calculation unit is combined with the input image by the image input unit by the image synthesis unit, the movement of the object for correcting the position determined by the graphic calculation unit is determined by motion information. The image processing apparatus according to claim 1, further comprising a motion detection unit for detecting.   The image processing apparatus according to claim 2, further comprising: a composition correction control unit that controls whether the image composition unit corrects a position where the composition target is synthesized with the input image. An image input step, an area detection step for detecting an object from an input image obtained by the image input step, a figure calculation step for determining a composition target and a position according to the object detected by the area detection step, and the image When synthesizing the synthesis target determined in the graphic calculation step with the input image in the input step, an image synthesis step for synthesizing the image based on the position determined in the graphic calculation step, and a composite image in the image synthesis step In the image processing method using a plurality of memories in order as a buffer for each step, while having an image output step for outputting,
In the first period, an input image is input to one memory in the image input step, and in the second period, the region detection step, the figure calculation step, and the image composition step are performed based on input information stored in the one memory. The image processing method is characterized in that the processing is executed in step (b) and the synthesized image obtained in the image synthesizing step is stored in the one memory that has been processed or a memory preceding the one memory.   When the synthesis target determined in the graphic calculation step is combined with the input image in the image input step in the image synthesis step, the movement of the object for correcting the position determined in the graphic calculation step is motion information. The image processing method according to claim 4, further comprising a motion detection step of detecting in step (a).   6. The image processing method according to claim 5, further comprising a composition correction control step for controlling whether or not to correct a position where the composition target is synthesized with the input image.   A mobile communication system comprising the image processing apparatus according to any one of claims 1 to 3.   An image editing system comprising the image processing apparatus according to any one of claims 1 to 3.   A program for causing a computer to execute the image processing method according to any one of claims 4 to 6.

Images