JP2005136841A - Device, method and program of image output processing, image distribution server and image distribution processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To combine and output a costume image with a pose corresponding to an object image according to a place and a season (time) about a photographed image of an image output device such as a digital camera in the image output device. <P>SOLUTION: An object figure image Hn for composition (imitation) is extracted about the photographed image 12in in accordance with performing photography, whether or not the current photography place is a specific place is determined, when the photography place is determined as the specific place, a face omitted costume image with the pose corresponding to the figure image Hn for composition is acquired with a costume of a type according to atmosphere of the place, a composite image 12in1 formed by replacing and combining the figure image Hn for composition with the acquired face omitted costume image according to the photography place is created and displayed. Thus, the composite image 12in1 in which costumes of an object figure are differentiated according to the atmosphere of a place where the photography is performed is photographed and displayed as the composite image 12in1 without a sense of incongruity by composition of the face omitted costume image with the pose imitating the pose the object figure image Hn. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image output device, an image output method, an image output processing program, an image distribution processing program, an image distribution processing program, and an image distribution processing program for synthesizing and outputting an image corresponding to a captured image in a digital camera, a camera-equipped mobile terminal, or the like. .

  2. Description of the Related Art Conventionally, there are various types of still cameras and video cameras that synthesize and display or print and output another image with a captured image.

  For example, an animation image such as a flower or a popular character prepared in advance at a fixed position of the shooting frame is output and synthesized, or an animation image prepared in advance is combined and output at a desired position in the shooting frame. Things are in practical use.

  In addition, an image captured by a digital camera can be displayed on a display screen of a personal computer and combined with another image for editing. In this case, for example, another image can be selectively selected from other image files. It is obtained by cutting out or downloading from any homepage screen on the Internet, and is synthesized at a desired position on the photographed image.

In addition, an arbitrary region (for example, a face portion) of an image photographed by a digital camera is cut out, and this cut-out image is combined with a prepared animation image (for example, the face portion of an anime character) to create a new another image. The animation image associated with the specified area of the image to be created or any image is synthesized (the cloud image is synthesized with the blue area estimated to be the blue sky, or the snow area is estimated to be the snow area) An image is synthesized (for example, see Patent Document 1).
JP 2001-230972 A

  In the conventional image processing system, an image prepared in advance is fixed to a captured image as it is or is combined and output at a desired position, or an image prepared in advance is combined with an arbitrary clipped image. The combined image other than the shot image is always fixed and does not change, and even if various images are downloaded from the Internet and used, they are closely related to each shot image. There is a problem that it is not easy to get tired of because it is just a decoration that is not related to.

  On the other hand, with the recent widespread use of small digital cameras and mobile phones with digital cameras, each user can always carry them around and take pictures anytime and anywhere.

  The present invention has been made in view of the above-described problems, and does not simply synthesize a fixed image with a captured image, but has a pose corresponding to the subject image according to the location and season (time). An object of the present invention is to provide an image output device, an image output method, an image output processing program, an image distribution server, and an image distribution processing program that can synthesize and output a costume image.

  The image output apparatus according to claim 1 of the present invention has a pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses in a type of costume according to position information, and is acquired by a captured image acquisition means. A person image to be synthesized is extracted from the captured image by a target image extraction unit, and position information of the captured image is acquired by a position information acquisition unit. Then, the costume image corresponding to the shooting position information of the shot image acquired by the position information acquisition unit and corresponding to the person image to be synthesized extracted by the target image extraction unit is the costume by pose. Acquired by the corresponding image acquisition means from the series of costume images stored by the image storage means. Then, the costume image acquired by the corresponding image acquisition unit is obtained by the image synthesis unit while leaving the face portion of the person image for the human image to be synthesized in the photographed image extracted by the target image extraction unit. The synthesized image obtained by synthesizing the costume image of the pose corresponding to the person image to be synthesized is output by the image output means.

  According to this, it is possible not only to acquire a costume image of a pose according to the person image to be synthesized in the photographed image and to synthesize and output only the face image and the person image to be synthesized, but also to shoot this photographed image. A composite image in which the person image on the photographed image is dressed in the costume according to the photographing position (location) as it is by compositing it with the person image to be synthesized as a type of costume image according to the position (location). Can be output as

  The image output apparatus according to claim 2 of the present invention has a pose-specific costume image storage means for storing a series of costume images composed of a plurality of poses in a type of costume according to time information, and is acquired by a captured image acquisition means. The person image to be synthesized is extracted from the captured image by the target image extraction unit, and the time information of the captured image is acquired by the time information acquisition unit. Then, a costume image corresponding to the shooting time information of the captured image acquired by the time information acquisition unit and in accordance with the person image to be synthesized extracted by the target image extraction unit is the costume by pose. Acquired by the corresponding image acquisition means from the series of costume images stored by the image storage means. Then, the costume image acquired by the corresponding image acquisition unit is obtained by the image synthesis unit while leaving the face portion of the person image for the human image to be synthesized in the photographed image extracted by the target image extraction unit. The synthesized image obtained by synthesizing the costume image of the pose corresponding to the person image to be synthesized is output by the image output means.

  According to this, it is possible not only to acquire a costume image of a pose according to the person image to be synthesized in the photographed image and to synthesize and output only the face image and the person image to be synthesized, but also to shoot this photographed image. A composite image in which the person image on the photographed image is changed into a costume according to the photographing time (period) by combining it with the person image to be synthesized as a costume image of a kind according to the time (period). Can be output as

  An image output apparatus according to a third aspect of the present invention is the image output apparatus according to the first or second aspect, wherein the target image extraction means is a series of a plurality of images acquired by the captured image acquisition means. A changed portion is extracted from the captured image as a person image to be synthesized.

  According to this, based on a series of a plurality of photographed images that are continuously sampled and captured at the time of photographing, the changed portion can be easily extracted as a person image to be synthesized.

  An image output device according to a fourth aspect of the present invention is the image output device according to the first or second aspect, wherein the target image extraction unit includes each of the captured images acquired by the captured image acquisition unit. A human image to be synthesized is extracted according to distance information between image portions.

  According to this, based on the shooting distance of each image portion of one shot image, the near image portion can be easily extracted as a person image of the person to be synthesized.

  An image output apparatus according to a fifth aspect of the present invention is the image output apparatus according to any one of the first to fourth aspects, wherein the corresponding image acquisition means extracts the image output apparatus by the target image extraction means. A similarity evaluation unit that evaluates the similarity between the person image to be synthesized in the captured image and the costume image for each pose stored by the pose-specific costume image storage unit, and is evaluated by the similarity evaluation unit. A costume image of a pose corresponding to the person image to be synthesized is acquired according to the similarity between the person image to be synthesized and each costume image.

  According to this, it is possible to easily acquire the type of costume image according to the shooting position (location) or shooting time (time) to be combined with the pose of the person image to be combined in the shot image.

  An image output apparatus according to a sixth aspect of the present invention is the image output apparatus according to any one of the first to fourth aspects, wherein the corresponding image acquisition means includes the pose-specific costume image storage means. A model image storage means for storing model images of a plurality of poses associated with each pose of the stored costume image; a person image to be synthesized in the photographed image extracted by the target image extraction means; and the model image storage means A degree-of-similarity evaluation unit that evaluates the degree of similarity of each pose stored in the model image according to the degree of similarity between the person image to be synthesized and the model image evaluated by the degree-of-similarity evaluation unit. A costume image associated with the pose of the model image and stored by the pose-specific costume image storage means is acquired.

  According to this, it is possible not only to easily acquire a type of costume image according to the shooting position (location) or shooting time (time) to be combined with the pose of the person image to be combined in the shot image, but also to obtain a similarity degree. Since a standard model is used for the evaluation object, it is easy to obtain a high degree of similarity (simulation) with the person image to be synthesized, and it is possible to acquire a costume image that is more comfortable in terms of expression.

  According to the image output apparatus of the first aspect of the present invention, a costume image in a pose corresponding to the person image to be combined in the photographed image is acquired and combined with the person image to be combined and the face portion remaining. In addition to outputting, the person image on the photographed image is captured in the pose as it is by synthesizing it with the person image to be synthesized as a costume image of a type corresponding to the photographing position (place) of the photographed image. It becomes possible to output as a composite image dressed in a costume according to.

  According to the image output apparatus of the second aspect of the present invention, a costume image in a pose corresponding to the person image to be combined in the photographed image is acquired and combined with the person image to be combined and the face portion remaining. In addition to outputting, the person image on the photographed image is captured in the pose as it is by synthesizing it with the person image to be synthesized as a costume image of a type corresponding to the photograph time (period) of the photographed image. It becomes possible to output as a composite image dressed in a costume according to.

  According to the image output device of the present invention, the image output device according to the first or second aspect of the present invention provides a series of a plurality of captured images that are continuously sampled and captured at the time of shooting. Based on this, the changed portion can be easily extracted as a person image to be synthesized.

  According to the image output device according to claim 4 of the present invention, in the image output device according to claim 1 or claim 2, based on the shooting distance of each image portion of one shot image, The front image portion can be easily extracted as a person image of the person to be synthesized.

  According to an image output device of a fifth aspect of the present invention, in the image output device according to any one of the first to fourth aspects of the present invention, imitating a pose of a person image to be synthesized in a photographed image. It is possible to easily acquire the type of costume image corresponding to the shooting position (location) or shooting time (time) to be synthesized.

  According to an image output apparatus according to claim 6 of the present invention, in the image output apparatus according to any one of claims 1 to 4, imitating a pose of a person image to be synthesized in a captured image. Not only can you easily acquire the type of costume image according to the shooting position (location) or shooting time (time) to be combined, but also use a standard model for the evaluation of similarity, so It is easy to obtain a high degree of similarity (imitation), and it becomes possible to acquire a costume image that is more comfortable in terms of expression.

  Therefore, according to the present invention, instead of simply synthesizing a fixed image with a photographed image, a costume image with a pose corresponding to the subject image is synthesized and output according to the location and season (time). Is possible.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram showing a configuration of an electronic circuit of the image output apparatus 10 according to the first embodiment of the present invention.

  The image output device 10 is configured as a portable device such as a digital camera, a mobile phone with a digital camera, or a PDA (personal digital assistants) with a digital camera. The image output device 10 is a program recorded on various recording media or transmitted. The electronic circuit is provided with a central processing unit (CPU) 11.

  The CPU 11 is a device control program stored in the memory 12 in advance, or a device control program read from the external recording medium 13 such as a CD-ROM into the memory 12 via the recording medium reading unit 14 or on the Internet N. The device stored in the memory 12 controls the operation of each part of the circuit in accordance with the device control program read into the memory 12 from the Web server 21 (in this case, the program server) via the power transmission control unit 15. The control program uses an input signal corresponding to a user operation from the input unit 16 including a key and a touch panel, an image signal input via the image input unit 17, or a position information detection unit (GPS: Global Positioning System) 18x. Current position information input via the interface, or an interface connected via the transmission control unit 15. It is activated in response to a communication signal with each Web server 21 on the network N.

  The CPU 11 is connected to the memory 12, the recording medium reading unit 14, the electric transmission control unit 15, the input unit 16, the image input unit 17, and the position information detection unit 18 x, and a display unit including a distance sensor 18 and an LCD. 19, an image output unit 20 and the like are connected.

  The image input unit 17 inputs captured image data, and the captured image data is selectively acquired from image capturing means such as a digital camera unit 17A and a digital video movie unit 17B. These image photographing means (17A) and (17B) are used separately from the main body of the image output apparatus 10 which is a portable device and connected by a connection cable, or combined and integrated by a connector. It is good also as a structure to do.

  The distance sensor 18 detects the distance to the subject of the captured image input by the image input unit 17.

  The image output unit 20 outputs image data processed by the present apparatus based on the captured image input from the image input unit 17, and the output image data includes a memory card 20A, a VTR (recording apparatus) 20B, The image is selectively output to the image printing machine 20C, and the image is stored, recorded, and printed.

  The memory 12 stores a system program that controls the overall operation of the image output apparatus 10 and a communication program for data communication with each Web server 21 on the Internet N via the transmission control unit 15. Replacement with the corresponding costume image (for example, various costume images imitating the pose of the subject person image) according to the synthesis target image (subject person image) in the photographed image via the image input unit 17 and the image output unit 20 ( An image output processing program 12a and the like for performing the dressing) composite output processing in association with the shooting location (position) and the date (time) are stored.

  The memory 12 also includes a model pose image memory 12b, a costume 1 “period drama” pose image memory 12c, a costume 2 “Western” pose image memory 12d, a costume 3 “medieval Europe” pose image memory 12e, and a costume 13 “ “Summer (June-September)” pose image memory 12f, Costume 14 “Spring and Fall (4, 5, 10, November)” pose image memory 12g, Costume 15 “Winter (December-March)” pose image memory 12h In addition, a captured image memory 12i, a composition target image position memory 12j, a position information registration memory 12k, a current date memory 12l, and a work area 12m are prepared.

  2 shows the costume 1 “period drama” pose image memory 12 c of the image output device 10: the costume 2 “Western” pose image memory 12 d... The costume 13 “summer (June-September)” pose image memory 12 f. It is a figure which shows the costume image 12cM, 12dM, and 12fM of each male version corresponding to various poses.

  FIG. 3 shows the costume 1 “period drama” pose image memory 12c: costume 2 “Western” pose image memory 12d... Costume 13 “summer (June to September)” pose image memory 12f of the image output device 10, respectively. It is a figure which shows the costume images 12cW, 12dW, and 12fW of each female version corresponding to various poses.

  Each costume image memory 12c, 12d, 12e, ... 13 costumes “Summer (June-September)”, costume 14 “Spring Autumn (4) , 5, 10 and November), costume 15 “winter (December to March)”, each pose image stored in each pose image memory 12f, 12g, 12h,. This is selectively used as a costume image that imitates the pose of the subject person image (image to be synthesized) on the photographed image input from the unit 17. Are stored in association with each other. In addition, when the pose image of each of these costumes is used as a costume image imitating the pose of the subject person image (image to be synthesized) on the photographed image, the location (position) and season ( The pose image of the type of costume specified according to the date is used.

  FIG. 4 is a view showing a model image corresponding to the model pose stored in the model pose image memory 12b of the image output apparatus 10. As shown in FIG.

  The model images of various poses stored in the model pose image memory 12b are similar model images corresponding to the pose of the subject person image (synthesize target image) on the photographed image input from the image input unit 17. This is used to acquire the pose No. for each different pose. Are stored in association with each other, and each pose No. is stored. Three types of model images of variations A, B, and C are stored every time.

  Then, the pose image memories 12c, 12d, 12e,... And the costumes 13 “Summer (June-September)”, 14 “Spring Autumn ( 4, 5, 10, November) ”, 15“ winter (December to March) ”, each costume image stored in each pose image memory 12 f, 12 g, 12 h,. All pose Nos. Stored in 12b. Costume images of the same pose corresponding to (No. 1A, 1B, 1C, 2A, 2B, 2C,..., NA, nB, nC) are prepared and stored.

  The costume images stored in the pose image memories 12c to 12h of the costumes are all the body parts below the hairstyle and neck, and the face part is removed to form a transparent image. By performing the replacement image composition by superimposing on the composition target image, a composite image in which only the face portion of the composition target image is changed and the costume is changed is created.

  The photographed image memory 12i stores photographed image data input via the image input unit 17 and composite image data created based on the photographed image data in units of photographed image units 1, 2,. In addition, each captured image data stored in the captured image memory 12i stores a plurality of captured preparation images sampled immediately before the captured image is captured (for example, when the shutter is half-pressed), or Shooting distance data in each part of the image area acquired along with AF (auto focus) processing of the shot image is stored together.

  In the compositing target image position memory 12j, the position information of each subject person image that is a compositing target extracted in each of the captured images 1, 2,... Stored in the captured image memory 12i is coordinated. It is stored in each photographed image unit 1, 2,.

  The location information registration memory 12k includes theme parks, event facilities, and amusement facilities for designating the pose images 12c, 12d, and 12e of the costumes 1 “period drama”, 2 “western”, and 3 “medieval Europe”. For example, information on a specific location (position) corresponding to each costume such as historical drama village, western village, euro village, etc. is registered in each location (position) unit 1, 2,.

  The current year / month / day memory 12l stores, for example, current year / month / day / time (time) data measured based on a clock signal output from the CPU 11.

  In the work area 12m, image data processed in accordance with the image output processing program 12a is temporarily stored according to the image processing.

  An outline of the composite image creation operation of the image output apparatus 10 configured as described above will be described.

  FIG. 5 is a diagram showing a replacement (dressing) composition state of a costume pose image corresponding to a photographing place (period play village) with respect to a person image to be synthesized (simulated) of the image output device 10.

  As shown in FIG. 5A, for example, in a historical village, when the photographed image 12i1 is captured from the image input unit 17 and displayed on the display unit 19 in response to a half-pressing operation of the shutter in the input unit 16, As shown in FIG. 5B, subject images (person images) H1, H2, and H3 are extracted from the photographed image 12i1 being photographed and displayed as shown in FIG. 5B. When it is determined that the current shooting position (location) detected by the position information detection unit 18x is a historical play village registered in the position information registration memory 12k, as shown in FIG. Similar (simulated) pose Nos. To the poses of the images (person images) H1, H2, and H3. The pose images 12cM (19), 12cW (19), and 12cM (19) of the 19th version “Squam” male version and female version of the period drama costume are read from the costume 1 “period drama” pose image memory 12c. Then, as shown in FIG. 5D, the costume 1 “period drama” pose image is placed on each of the subject images (person images) H1, H2, and H3 to be synthesized (simulated) on the photographed image 12i1. The pose No. read from the memory (male version / female version) 12c. Each pose image 12cM (19), 12cW (19), 12cM (19) in 19 drama style historical drama style is synthesized by replacing only the face portion of the original photographed image 12i1.

  As a result, when taking a picture in a historical drama village registered in the position information registration memory 12k, a “drama play” style costume imitating the pose of each subject (person image) H1, H2, H3 in the shot image 12i1. The pose images 12cM (19), 12cW (19), and 12cM (19) are replaced and combined with the respective subjects (person images) H1, H2, and H3, leaving only the face portions, and the costumes of the original subject images H1, H2, and H3 are combined. It becomes possible to get the display by changing into a period drama costume.

  The distinction between male / female in each subject person image H1, H2, H3 extracted from the photographed image 12i1 is specified and distinguished by the user on the extraction display screen of each subject person image H1, H2, H3. 1 “period drama” pose image memory (male version: see FIG. 2) 12cM (19) of a similar pose is read from 12c or the same costume 1 “period drama” pose image memory (female version: see FIG. 3) ) A female costume image 12cW (19) in a similar pose is read from 12c. Also, if the user does not distinguish between male / female on the extraction display screen of the subject person images H1, H2, H3, the costume 1 “period drama” pose image memory corresponding to the location information (male version) / Women version) The male or female costume pose image 12cM (19) or 12cW (19) having the higher similarity is read out from 12c.

  FIG. 6 is a diagram showing a replacement (dressing) composition state of a costume pose image according to a photographing location (Western village) for a person image to be synthesized (simulated) of the photographed image of the image output device 10.

  As shown in FIG. 6A, for example, in Western village, when the photographed image 12i2 is captured from the image input unit 17 and displayed on the display unit 19 in accordance with the half-pressing operation of the shutter in the input unit 16, this photographing is performed. As shown in FIG. 6B, subject images (person images) H1, H2, and H3 are extracted from the captured image 12i2 being displayed. When it is determined that the current shooting position (location) detected by the position information detection unit 18x is a western village registered in the position information registration memory 12k, as shown in FIG. (Portrait image) A pose No. similar to (similar to) each pose of H1, H2, and H3. 4 “Walking (horizontal) inversion” male version Western style costume pose image 12dM (4) R, pose no. 20 “Banzai” male version Western style costume pose image 12dM (20), pose no. 4 “Walking (horizontal)” female version Western style costume pose image 12dW (4) is read from the costume 2 “Western” pose image memory 12d. Then, as shown in FIG. 6D, the costume 2 “Western” pose image memory is located at each of the subject images (person images) H1, H2, and H3 on the photographed image 12i2. (Male version / Female version) Pose No. read from 12d 4 “Walking (horizontal) reversal”, no. 20 “Banzai”, no. 4 Western style pose images 12dM (4) R, 12dM (20), and 12dW (4) of “walking (horizontal)” are replaced and replaced with only the face portion of the photographic source image 12i1.

  As a result, when shooting in the Western village registered in the position information registration memory 12k, a “Western” style costume pose image that imitates the pose of each subject (person image) H1, H2, H3 in the captured image 12i2. 12dM (4) R, 12dM (20), and 12dW (4) are replaced and combined with the respective subjects (person images) H1, H2, and H3, leaving only the face portions, and the costumes of the original subject images H1, H2, and H3 are combined. You can get the display by changing into a Western costume.

  In this case as well, the distinction between the male version and female version of the costume 2 “Western” pose image (12d) corresponding to each subject person image H1, H2, H3 is the same as in the case of the costume 1 “period drama”. Is done. Hereinafter, it is performed in the same manner in other cases.

  FIG. 7 is a diagram showing a replacement (dressing) composition state of a costume pose image according to a photographing season (year / month / day / time) with respect to a person image to be synthesized (simulated) of the image output device 10.

  As shown in FIG. 7A, when the photographed image 12i3 is captured from the image input unit 17 and displayed on the display unit 19 in accordance with the half-pressing operation of the shutter in the input unit 16, the photographed image being photographed and displayed is displayed. From 12i3, as shown in FIG. 7B, a subject image (person image) H1 as a synthesis target is extracted. When it is determined that the current date updated and stored in the current date memory 12l is August in summer, as shown in FIG. 7C, the subject image (person image) H1 is put into a pose. Similar (imitation) pose No. The summer pose image 12fW (1) of the female version of 1 “yes cheese (up)” is read from the costume 13 “summer” pose image memory 12f. Then, as shown in FIG. 7D, from the costume 13 “summer” pose image memory (female version) 12f at the position of the subject image (person image) H1 to be synthesized (simulated) on the photographed image 12i3. The read pose No. A summer swimsuit-style pose image 12fW (1) of 1 “yes cheese (up)” is replaced and replaced with the face image of the original image 12i3.

  As a result, when the image is taken in August in summer, the “summer swimsuit” style costume pose image 12fW (1) simulating the pose of the subject (person image) H1 in the photographed image 12i3 is the subject (person image). Only H1 and the face part are left and replaced, and the costume of the original subject image H1 can be changed into a summer costume and displayed and acquired.

  That is, for each photographed image 12in, when a subject person image Hn to be synthesized (simulated) is extracted, a costume pose imitating the pose of the subject image Hn with a kind of costume according to the photographing location (position) and the photographing season. The image is replaced and synthesized by leaving only the subject image Hn and the face portion thereof, and even if the person who is the subject of photography does not wear a costume according to the photographing location and photographing time, the location and season It is possible to capture and acquire a subject image changed to a costume corresponding to the.

  Next, a detailed image processing operation by the image output apparatus 10 having the above configuration will be described.

  Here, a case where the image output apparatus 10 is implemented as a digital camera will be described.

  FIG. 8 is a diagram showing an external configuration of the digital camera 10 as the image output device and its cradle 15x.

  The input unit 16 of the digital camera 10 includes a “MENU” key 16a for displaying a menu screen, a switching key 16b for switching between a shooting mode (REC) and a playback mode (PLAY), and capturing and recording of a shot image. There are provided a shutter key 16c for instructing and a selection / decision key 16d operable with one finger up, down, left and right for operations such as data selection on the display screen (19), cursor movement and selection data determination. .

  The digital camera 10 can be easily attached to and detached from a cradle 15x serving as a communication interface with a charging base of the camera 10 and a user PC (personal computer). When the “PHOTO” button 15a is pressed while the digital camera 10 is mounted on the cradle 15x, each captured image 12in stored in the captured image memory 12i and its composite image 12inm are read out from the display unit 19. Display (photo stand function) is performed.

  When the “USB” button 15b is pressed, each photographic image 12in stored in the photographic image memory 12i and its composite image 12inm are transferred and output to the user PC.

  The shutter key 16c is a key for instructing capture of a captured image in two steps of half-press and full-press. In the half-press state, the captured image 12in input from the image input unit 17 at a constant sampling interval is AE ( Auto exposure), AF (auto focus), and the like are adjusted and repeatedly written in the image processing work area 12m, displayed on the display unit 19, and registered in the photographed image memory 12i in the fully depressed state.

  FIG. 9 is a flowchart showing image composition processing of the image output apparatus 10.

  When shooting an image, if the shutter key 16c of the input unit 16 is pressed halfway, the input of the captured image 12in from the image input unit 17 is started, written in the image processing work area 12m, and displayed on the display unit 19 in real time. (Step S1).

  Then, the current shooting position (location) is detected by the position information detection unit 18x (step S2), and whether or not it matches the position information of the specific location registered in the position information registration memory 12k, that is, the current It is determined whether or not the shooting location (position) is a specific location such as a historical village or a Western village (step S3).

  Here, if it is determined that the current shooting location (position) matches any of the specific locations registered in the location information registration memory 12k, the type of costume image corresponding to the current location information, That is, if the shooting location is, for example, a historical drama village, the costume 1 “period drama” pose image (12c) is set as the type of costume image to be synthesized, and if the shooting location is, for example, a Western village, The costume 2 “Western” pose image (12d) is set as the type of costume image to be synthesized (steps S3 → S4).

  Then, the process proceeds to extraction processing of the synthesis (simulation) target image in FIG. 10 (step SA).

  On the other hand, if it is determined that the current shooting location (position) does not match any of the specific locations registered in the location information registration memory 12k, the current year updated and stored in the current date memory 12l. The month is detected from the date / time information (step S3 → S5).

  Then, if the type of costume image corresponding to the current month, that is, the current month is, for example, August in midsummer, the costume image to be synthesized with the costume 13 “Summer (June to September)” pose image (12f). If the current month is, for example, May in spring, the costume 14 “Spring Autumn (4, 5, 10 and November)” pose image (12d) is set as the type of costume image to be synthesized. (Step S6).

  Then, the process proceeds to extraction processing of the synthesis (simulation) target image in FIG. 10 (step SA).

  FIG. 10 is a flowchart showing a target image extraction process for extracting a person image to be synthesized in accordance with the image synthesis process of the image output apparatus 10, and FIG. 10A shows the target image extraction process (method 1). FIG. 5B is a flowchart showing the target image extraction process (method 2).

  In the target image extraction process (method 1) in FIG. 10A, a series of photographed images 12inm... Sequentially input at a predetermined timing via the image input unit 17 is stored in the photographed image memory 12i (step S1). A11).

  Then, from each of the series of photographed images 12 inm... Stored in the photographed image memory 12 i, the changed portion of the image accompanying the movement of the subject is extracted for each block of the subject (step A 12), and each of the extracted parts is thereby extracted. The subject image Hn is stored in the image processing work area 12m for each block (step A13).

  In the target image extraction process (method 2) in FIG. 10B, distance measurement is performed by the distance sensor 18 for each part of the captured image 12inm input via the image input unit 17 (step A21), and the background Each subject image Hn, which is discriminated as a short distance side based on the distance difference between the image and the subject image, is extracted for each block (step A22).

  Each subject image Hn thus extracted is stored in the image processing work area 12m for each block (step A23).

  Thus, when each subject image Hn included in the photographed image 12inm is extracted for each block from the photographed image 12inm and stored in the image processing work area 12m (step SA), for example, FIG. 5A or FIG. As shown in FIG. 7A or FIG. 7A, a photographed image 12in in which a surrounding mark M is added to each of the extracted subject person images Hn is displayed on the display unit 19 and is displayed on the photographed image 12in. The position information of each subject person image Hn is stored in the synthesis target image position memory 12j.

  When the subject person image Hn to be synthesized (simulated) is extracted and set from the photographed image 12in, the process proceeds to the corresponding costume image acquisition process in FIG. 11 (step SB).

  FIG. 11 is a flowchart showing a corresponding costume image acquisition process associated with the image composition process of the image output apparatus 10, and FIG. 11A is a flowchart showing the corresponding costume image acquisition process (method 1), and FIG. Is a flowchart showing the corresponding costume image acquisition processing (method 2).

  In the corresponding costume image acquisition process (method 1) in FIG. 11A, first, the subject person image Hn extracted from the photographed image 12in and set as a synthesis target image and the photographing position (location) or photographing month. The degree of similarity between each pose image of the type of costume set according to (year / month / day / time) is evaluated (step B1C), and the currently photographed image 12in is displayed as the costume pose image having the highest similarity. For example, a window screen is displayed on the display unit 19 (step B12).

  Here, when a costume pose image having the highest similarity to the synthesis target image displayed on the display unit 19 in response to a user operation on the input unit 16 is selected, the selected pose No. is selected. Of costume images are registered as costume images of poses corresponding to the synthesis target person image Hn (step B13 → B14).

  On the other hand, “next display” is instructed in response to a user operation on the input unit 16 in a state where a costume pose image having a pose with the highest degree of similarity is displayed on the display unit 19 as a synthesis target person image Hn. Then, for each “next display” operation, the costume pose image having the next highest similarity is displayed in order (step B13 → B15).

  Then, a costume pose image having a desired similarity pose is selected and displayed. When the displayed costume pose image is selected in response to a user operation on the input unit 16, the selected pose No. Of costume images are registered as costume images of poses corresponding to the synthesis target person image Hn (step B13 → B14).

  In this way, in the corresponding costume image acquisition process (method 1), the shooting position (location) or shooting month (year / month / day) is applied to the subject person image Hn as the synthesis target image extracted and set from the shot image 12in. The degree of similarity of each type of costume set according to the time) is directly evaluated, and based on this, the costume pose image to be replaced (changed) by imitating the pose of the subject person image Hn is determined.

  In the corresponding costume image acquisition process (method 2) in FIG. 11B, first, the subject person image Hn extracted from the photographed image 12in and set as a synthesis target image and stored in the model pose image memory 12b. The degree of similarity of each pose with the model image is evaluated (step B2C), and the pose No. of the model image of the pose with the highest degree of similarity is evaluated. A set-type clothing image corresponding to the shooting location or shooting month is displayed on the display unit 19 on which the current shot image 12in is displayed, for example, as a window screen (step B22).

  Here, when a costume pose image corresponding to the model pose having the highest similarity is selected in the imitation target person image Hn displayed on the display unit 19 in response to a user operation on the input unit 16, the selected pose is selected. No. Costume pose images are registered as costume images of poses corresponding to the synthesis target person image Hn (steps B23 → B24).

  On the other hand, the pose No. of the model image of the pose having the highest similarity to the target human image Hn is displayed on the display unit 19. When a “next display” is instructed in response to a user operation on the input unit 16 in a state where a set-type costume image corresponding to the shooting location or shooting month is displayed, the operation of the “next display” is performed. The pose No. of the model image with the next highest similarity for each time. Are displayed in order (step B23 → B25).

  Then, a set-type costume image corresponding to a model pose having a desired similarity is selected and displayed, and when the displayed costume pose image is selected in response to a user operation on the input unit 16, the selected pose No. . Costume pose images are registered as costume images of poses corresponding to the synthesis target person image Hn (steps B23 → B24).

  As described above, in the corresponding costume image acquisition process (method 2), the subject person image Hn as the imitation target image extracted and set from the captured image 12in is similar to the model image for each pose that is a basic model of the person. The degree is evaluated, and based on this, a costume pose image to be replaced (changed) by imitating the pose of the subject person image Hn is determined.

  In this case, for the subject person image Hn as the imitation target image, the costume image for replacement (dressing) is determined by directly evaluating the degree of similarity with each pose image of the set type according to the shooting location or shooting month. Compared with the corresponding costume image acquisition processing (method 1), the degree of similarity with the model image for each pose is evaluated, and the selected model pose No. Since the corresponding costume image acquisition process (method 2) in which the costume image to be replaced (changed) is determined from a simple person model without decoration as the evaluation target of the similarity, the subject person image Hn It is easy to obtain a high degree of similarity, and there is an advantage that a costume image of a pose with higher similarity (similarity) can be acquired.

  FIG. 12 is a flowchart showing similarity evaluation processing with the person image to be imitated in the corresponding costume image acquisition processing (method 1) (method 2) accompanying the image composition processing of the image output apparatus 10.

  In the case of the similarity evaluation process in the corresponding costume image acquisition process (method 1), the similarity with the pose image of the set type of costume is evaluated, and the similarity evaluation process in the corresponding costume image acquisition process (method 2) In this case, the similarity with the model image for each pose is evaluated.

  FIG. 13 is a diagram illustrating an example of similarity evaluation with a set-type costume pose image associated with the similarity evaluation process in the corresponding costume image acquisition process (method 1) of the image output apparatus 10.

  FIG. 14 is a diagram showing an example of similarity evaluation with a model image accompanying the similarity evaluation process in the corresponding costume image acquisition process (method 2) of the image output apparatus 10.

  First, in the case of similarity evaluation processing (method 1), for example, as shown in FIG. 13A, the first pose No. in the set type of costume (for example, costume 13 “Summer (June-September)” 12f). Costume image 12fW (1) is read out (step C1), and the vertical and horizontal sizes of the imitation target person image H1 shown in FIG.

  Then, a superimposition evaluation of the imitation target image H1 and the summer costume pose image 12fW (1) in which the vertical and horizontal sizes are combined is performed, and the evaluation value (%) is calculated and the summer costume pose image 12fW (1) is calculated. (Step C3).

  Here, the pose No. of the similarity evaluation target this time. It is determined whether or not the costume pose image 12fW (1) is a left-right asymmetric image (step C4). If it is determined that the costume image in the pose is not a left-right asymmetric image (step C4 “No”). ), The next pose No. The similarity calculation and storage processing by the size matching processing and the overlay evaluation value calculation processing are repeated with the summer costume image 12fW (2)... (Step C8 → C9 → C2 to C4).

  On the other hand, for example, as shown in FIG. If it is determined that the summer costume image is a left-right asymmetric photographer image (for example, “Goodbye” 12 fW (11)), for example, as shown in FIG. 13C, the costume image (12 fW ( 11)) is reversed left and right (step C4 → C5), and the vertical and horizontal sizes of the reversed costume image (12fW (11) R) and the imitation target image are matched (step C6).

  Then, the overlay evaluation of the imitation target image and the inverted costume image in which the vertical and horizontal sizes are combined is performed, and the evaluation value (%) is calculated and stored as the similarity with the inverted summer costume pose image (step) C7).

  After this, the next pose No. in the set-type costume is displayed. If there is a costume image (for example, 12 fW (n)) (step C8), the next pose No. The costume images (12fW (n)) are sequentially read out (step C9), and the similarity evaluation process between the person image Hn to be imitated and the costume image (12fW (n)) for each set pose is performed as described above. Repeated (steps C2 to C7).

  Thus, for example, as shown in FIG. 13, when the similarity evaluation with the “summer” costume image 12fW (n)... . The summer costume image 12fW (1) corresponding to 1 is evaluated as the costume pose image with the highest similarity.

  Next, in the similarity evaluation process (method 2), for example, as shown in FIG. Model image 12b (1A) is read out (step C1), and the vertical and horizontal sizes of the imitation target person image H1 shown in FIG.

  Then, the overlay evaluation of the imitation target person image H1 and the model image 12b (1A) in which the vertical and horizontal sizes are combined is performed, and the evaluation value (%) is calculated to obtain the model image 12b (1A) of the pose. The similarity is stored (step C3).

  Here, the pose No. of the similarity evaluation target this time. It is determined whether or not the model image is a left-right asymmetric image (step C4). For example, as shown in FIG. 14A, it is determined that the model image 12b (1A) of the pose is not a left-right asymmetric image. If it is (step C4 “No”), the next pose No. Model image is read out (steps C8 → C9), and similarly, the similarity evaluation process between the imitation target person image H1 and the model image of the next pose is repeated (steps C2 to C4).

  On the other hand, for example, as shown in FIG. When it is determined that the 1B model image 12b (1B) is a left-right asymmetric model image, the model image 12b (1B) of the pose 1B is reversed left and right (step C4 → C5), and FIG. As shown in FIG. 6, the vertical and horizontal sizes of the inverted model image 12b (1B) R and the imitation target person image H1 are matched (step C6).

  Then, the overlay evaluation of the imitation target person image H1 and the inverted model image 12b (1B) R in which the vertical and horizontal sizes are combined is performed, and the evaluation value (%) is calculated and the inverted model image 12b (1B) of the pose is calculated. ) Stored as the similarity to R (step C7).

  After this, the next pose No. If there is a model image 12b (n) (step C8), the next pose No. Model images 12b (n) are sequentially read (step C9), and similarly, the similarity evaluation process between the imitation target person image Hn and the model image 12b (n) for each pose is repeated (steps C2 to C7). ).

  As a result, for example, as shown in FIG. 14, when the similarity evaluation with the model image 12b (n) is performed on the subject person image H1 to be synthesized (simulated), the pose No. The model image 12b (1A) corresponding to 1A is evaluated as a model image with the highest similarity.

  As described above, by the corresponding costume image acquisition process (step SB) described with reference to FIGS. 11 to 14, the pose setting type imitating the subject person image Hn as the imitation target image extracted from the photographed image 12 in. When the costume image is determined, for example, as shown in FIG. 5, the image to be imitated on the photographed image 12 i 1 acquired by the corresponding costume image acquisition process (see FIG. 11) in the shooting at the historical drama village. The costume pose images 12cM (19), 12cW (19), and 12cM (19) of “period drama”, which is a set of costumes that imitate each pose “squatting” of the subject person H1, H2, and H3, Each of the subject person images H1, H2, and H3 to be synthesized (simulated) is replaced (changed) while leaving only its face portion, and is synthesized and displayed (steps S7 and S8).

  In this case, the historical drama costume images 12cM (19), 12cW (19), and 12cM (19) of the poses similar to the subject person images H1, H2, and H3 to be synthesized (simulated) with respect to the photographed image 12i1. Are superimposed on the positions of the subject images H1, H2, and H3 (composite target image positions 12j1) in order of increasing shooting distance detected by the distance sensor 18 at the time of shooting.

  That is, since there is no image in the face portion of each costume image, the face portions of the subject person images H1, H2, H3 in the photographed image 12i1 are left and the historical drama costume images 12cM (19), 12cW (19 ), 12cM (19) are replaced and the costume of each subject person image H1, H2, H3 is changed to the period drama style with respect to the original photographed image 12i1 in the period drama village. A composite image 12i11 is generated and displayed.

  Then, as shown in FIG. 5D, the historical drama costume images 12cM (19), 12cW (19), 12cM (19) imitating the poses of the subject images H1, H2, H3 are the subject image H1. , H2, H3 and the combined image data 12i11 that is replaced by leaving only its face portion is registered in the photographed image memory 12i when the full press operation of the shutter key 16c is detected (step S9).

  For example, as shown in FIG. 6, in the shooting in the Western village, the subject persons H1, H2, and H3, which are imitation target images on the captured image 12i2, acquired by the corresponding costume image acquisition process (step SB). Each pose image of “Western”, which is a set of costumes imitating “walking (horizontal) reversal”, “banzai”, and “walking (horizontal)” 12dM (4) R, 12dM (20), 12dW ( 4) are respectively replaced (dressed) with the subject person images H1, H2, H3 and their face portions to be combined (simulated) displayed as combined (steps S7, S8).

  In this case, the western costume images 12dM (4) R, 12dM (20), and 12dW (4) having similar poses to the subject person images H1, H2, and H3 to be synthesized (simulated) with respect to the photographed image 12i2. Are superimposed on the positions of the subject images H1, H2, and H3 (synthesize target image positions 12j2) in order of increasing shooting distance detected by the distance sensor 18 at the time of shooting.

  That is, also in this case, since there is no image in the face portion of each costume image, the face portions of the subject person images H1, H2, H3 in the photographed image 12i2 are left and the Western costume images 12dM (4) R, Since 12dM (20) and 12dW (4) are superposed and replaced, in the shooting at the Western village, the costumes of the subject person images H1, H2 and H3 are changed to the Western style with respect to the original shot image 12i2. A composite image 12i21 is created and displayed.

  As shown in FIG. 6D, the Western costume images 12dM (4) R, 12dM (20), and 12dW (4) imitating the poses of the subject images H1, H2, and H3 are the subject image H1. , H2 and H3 and the combined image data 12i21, which is replaced by leaving only its face portion, is registered in the photographed image memory 12i when the full depression of the shutter key 16c is detected (step S9).

  On the other hand, for example, as shown in FIG. 7, in August shooting when the position information of the shooting location does not match the specific position information, the shot image 12i3 acquired by the corresponding costume image acquisition processing (step SB). The costume pose image 12fW (1) of “Summer”, which is a set of costumes that imitates the pose “yes cheese (up)” of the subject person H1 that is the image to be imitated above, is the subject person to be synthesized (simulated). Only the image H1 and its face part are left and replaced (dressed) and combined and displayed (steps S7 and S8).

  In this case, with respect to the photographed image 12i3, the summer costume image 12fW (1) having a pose similar to that of the subject person image H1 to be synthesized (simulated) is set to the position of the subject image H1 (synthesized target image position 12j3). Overlay and synthesize.

  That is, also in this case, since there is no image in the face portion of the costume image, the face portion of the subject person image H1 in the photographed image 12i3 is left and replaced with the summer costume image 12fW (1). In shooting in August, a composite image 12i31 is created and displayed by changing the costume of the subject person image H1 to the swimsuit style with respect to the original shot image 12i3.

  Then, as shown in FIG. 7D, the composite image data 12i31 obtained by replacing the summer costume image 12fW (1) imitating the pose of the subject image H1 while leaving only the subject image H1 and its face portion is obtained. When the full depression of the shutter key 16c is detected, it is registered in the photographed image memory 12i (step S9).

  The photographed (synthesized) image registered in the photographed image memory 12i is output from the image output unit 20 as necessary, and recorded by the memory card 20A, recorded by the VTR (recording device) 20B, or image printing machine 20C. Is printed.

  As a result, when photographing at a specific location registered in the position information registration memory 12k, the subject person image Hn on the photographed image 12in is photographed and displayed as a costume image corresponding to the atmosphere of the photographing location. In addition, the current shooting month is detected at a place other than a specific place, and the subject person image Hn on the shot image 12in is changed to a costume image corresponding to the season of the shooting month and is displayed. Become so.

  Therefore, according to the image composition processing function of the image output apparatus 10 having the above-described configuration, when the subject person image Hn to be synthesized (simulated) is extracted from the photographed image 12in as the photograph is taken, the current photographing location is obtained. It is determined whether or not (position) is a specific place, and if it is determined that the position is a specific place, it corresponds to the person image Hn to be synthesized (simulated) with a kind of costume according to the atmosphere of the place. A face-out costume image of the pose to be acquired is acquired, and a composite image 12in1 is created and displayed by replacing the composition target person image Hn with the face-out costume image corresponding to the acquired shooting location. Therefore, the composite image 12in1 in which the subject person's costume is changed in accordance with the atmosphere of the place where the photograph is taken is combined with the face-out costume image in a pose imitating the pose of the subject person image Hn. Very becomes possible shooting display as a composite image 12in1 with interesting without discomfort by.

  Further, according to the image composition processing function of the image output apparatus 10 having the above configuration, when the subject person image Hn to be synthesized (simulated) is extracted from the photographed image 12in as the photograph is taken, the current photographing time is obtained. (Season / Year / Month / Day / Hour) is detected, and, for example, a faceless outfit image of a pose corresponding to the synthesis target person image Hn is acquired with a type of costume adapted to the atmosphere of the season. The composite image 12inm is created and displayed by replacing the synthetic target person image Hn with the face-out costume image corresponding to the acquired shooting season, so that it matches the atmosphere of the season in which the photo is taken. A composite image 1 in which the costume of the subject person is made different from the composite image 12 inm with a face-out costume image in a pose that imitates the pose of the subject person image Hn is very interesting and comfortable. It will be able to shooting display as inm.

  Further, according to the target image extraction function (method 1) accompanying the image composition processing by the image output apparatus 10 having the above-described configuration, a series of captured images 12in... Sequentially input at a predetermined timing via the image input unit 17 is an image. Since the change part of the image accompanying the movement of the subject is extracted from each of the series of photographed images 12in... Stored in the processing work area for each block M of each subject person image, a series of images including the photographed images 12in is included. The image portion of each subject person image Hn... Extracted based on the captured image 12in... Is easily combined (simulated) target image position memory 12j as the position information of the combined target image Hn associated with the captured image 12in. Can be memorized.

  Further, according to the target image extraction function (method 2) accompanying the image composition processing by the image output apparatus 10 having the above configuration, the distance by the distance sensor 18 for each part of the captured image 12in input via the image input unit 17. Each subject person image Hn, which is measured and stored in the image processing work area and is determined as the near distance side based on the distance difference between the background image and the subject image, is extracted for each block M. The image portion of each of the subject person images Hn... Extracted based on the one captured image is easily combined (simulated) target image position as position information of the combined target image Hn associated with the captured image 12in. It can be stored in the memory 12j.

  Further, according to the corresponding costume image acquisition function (method 1) accompanying the image composition processing by the image output apparatus 10 having the above-described configuration, the shooting location and the subject person image Hn as the imitation target image extracted from the captured image 12in The degree of similarity between the type of costume set according to the shooting time and the faceless outfit image for each pose is directly evaluated, and based on this, the set type of costume image that imitates the pose of the subject person image Hn is determined. Therefore, it is possible to easily acquire a costume image for imposing a pose of the subject person image Hn and replacing (changing) the original image with the image of the photographic costume.

  Further, according to the corresponding costume image acquisition function (method 2) accompanying the image composition processing by the image output apparatus 10 having the above-described configuration, the person basic is applied to the subject person image Hn as the imitation target image extracted from the photographed image 12in. The degree of similarity with the model image for each pose that is a model is evaluated, and based on this, a set type of costume image is determined according to the shooting location and shooting time in a pose that imitates the subject person image Hn. It is possible not only to easily acquire a costume image for imposing a pose of the image Hn and replacing (changing) with the original image of the clothing costume part, but also to use a standard person model as the evaluation target of similarity. It is easy to obtain a high degree of similarity with the person image Hn, and it is possible to acquire a costume image in a pose with a higher degree of similarity (similarity).

  In the image composition function in the above embodiment, a face-out costume image of a set type of pose corresponding to the subject person image Hn extracted from the photographed image 12in is selectively acquired, and the acquired face-out costume image is acquired. Is replaced with the subject person image Hn of the photographed image 12in, but various kinds of costume images based on live-action and computer graphics close to real-life are stored in advance for each pose, and each pose by this live-action etc. The costume image may be set as the type of costume image according to the place and season, and replaced with a costume image corresponding to the pose of the subject person image Hn extracted from the photographed image 12in and synthesized.

  In the embodiment, the case where the captured image 12in is a still image has been described. However, based on a moving image captured and input by, for example, the digital video movie 17B via the image input unit 17, a background in the moving image is displayed. A subject image Hn that changes in time series with respect to the image is sequentially extracted, a costume image of a pose setting type corresponding to the extracted subject person image Hn is acquired, and the subject in the captured moving image is acquired. It may be configured to be replaced and synthesized with the person image Hn.

  In the image output device 10 of the above-described embodiment, the type of face is determined according to the shooting location and shooting time of the pose imitating the subject person image Hn in the shot image 12in in the mobile phone with digital camera or the main body of the digital camera. The costume image is acquired and the costume image that replaces only the face portion of the subject person image Hn is replaced (changed). However, as described in the second embodiment, the Internet or the like is used. The Web server 30 on the communication network N receives a photographed image accompanying a photograph taken from each access user, and the subject person image Hn to be synthesized (simulated) is extracted from the received photographed image in the same manner as described above. , A process of acquiring a face-out costume image of a type corresponding to the shooting location and shooting time in a pose imitating the subject person image Hn, and Performs replace composition process of the face vent costume image for subject person image Hn of the received captured image may be configured to deliver the synthesized image to the access source user terminal.

(Second embodiment)
FIG. 15 is a block diagram showing a configuration of an electronic circuit of the image distribution server 30 according to the second embodiment of the present invention.

  The image distribution server 30 is configured by a computer that reads a program recorded on various recording media or a transmitted program and whose operation is controlled by the read program. The electronic circuit includes a CPU ( a central processing unit) 31 is provided.

  The CPU 31 is a server control program stored in advance in the memory 32, or a server control program read from the external recording medium 33 such as a CD-ROM into the memory 32 via the recording medium reading unit 34, or on the Internet N. The operation of each part of the circuit is controlled in accordance with a server control program read into the memory 32 from another Web server (in this case, a program server) via the power transmission control unit 35 and stored in the memory 32. The server control program includes communication terminals (mobile phone with digital camera 10A, digital camera with communication function 10B, other PC (personal computer) terminal) on the Internet N connected via the transmission control unit 35, and other Web It is activated in response to a communication signal with the server.

  In addition to the memory 32, the recording medium reading unit 34, and the power transmission control unit 35, the CPU 31 is connected to a display unit 36 including an LCD.

  The memory 32 stores a system program that controls the overall operation of the image distribution server 30 and each communication terminal (mobile phone 10A with a digital camera, digital camera 10B with a communication function) on the Internet N via a transmission control unit 35. In addition to storing communication programs for data communication with other Web servers, other PC (personal computer) terminals, etc., based on access to each communication terminal (10A, 10 etc.) on the Internet N, Each of the user terminals generates and manages captured image files 32g1, 32g2,..., Stores and distributes captured image data, stores synthesis target image positions and captured position information about the captured images, An image output processing program for creating a composite image for a shot image and for outputting and delivering the composite image 32a, etc. are stored.

  Accordingly, in the memory 32 of the image distribution server 30, the model pose image memory 32b, the costume 1 “period drama” pose image memory 32c, and the costume 2 “Western” pose image are substantially the same as the memory 12 of the image output apparatus 10. Memory 32d, costume 3 "Medieval Europe" pose image memory 32e ... and costume 13 "summer (June-September)" pose image memory not shown, costume 14 "Spring fall (4, 5, 10, November)" pose image Memory, costume 15 “winter (December to March)” pose image memory, etc. are prepared, and in addition to the current date memory 32f, the work distribution area 30 takes a picture of each user terminal. Shooting image files 32g1, 32g2,... For managing image data and setting information for shooting image composition are prepared.

  The captured image file (1) 32g1, (2) 32g2,... For each user terminal includes a user name: file name: user authentication data (fingerprint image or password) for each of the files (1), (2),. , Etc.) 32g1a, 32g2a,..., Image image memory 32g1b, 32g2b,..., Composition target image position memory 32g1c, 32g2c,. The image memory 32g1b, 32g2b,... And the synthesis target image position memory 32g1c, 32g2c,... Store and manage the same information as each memory in the image output device 10 of the first embodiment, and capture position information. In the memories 32g1d, 32g2d,..., Position information 32g1dn, 32g2dn indicating the shooting locations of the respective captured images 32g1bn, 32g2bn,. ... are stored and managed are received together.

  The image distribution server 30 having such a configuration is established on a communication network N such as the Internet, and communicates with the mobile phone 10A with a digital camera or the digital camera 10B with a communication function as each user terminal. The location information of the shooting location is received together with the shot image, and is managed in the image distribution server 30 in the same manner as the image output device 10 of the first embodiment. Thus, the image distribution server 30 receives a captured image with shooting position information from each user terminal, creates a composite image based on the received image, and requests for distribution of the received captured image. While extracting the person image Hn, it is possible to acquire a faceless costume image in a pose imitating the subject person image Hn with a type of costume image according to the shooting location or shooting time, and the image output apparatus 10 of the first embodiment. Similarly to the above, it is possible to create a composite image in which the subject person image Hn is replaced with a faceless costume image. Then, the composite image can be distributed to each user terminal as a distribution request source.

  In this case, the composite image creation process and the distribution process based on the received photographed image and the photographing position information in each photographed image file 32gn are performed when the authentication data 32gna of each corresponding user terminal is determined to match the input. Only run on.

  Therefore, according to the image synthesizing function and the synthesized image distribution function performed while accessing the respective user terminals in the image distribution server 30 having the above-described configuration, the captured image 32gnb and the captured position information 32gnd are received from each user terminal. In addition to receiving the composite image creation and distribution request, in response to the composite image creation and distribution request, whether the photographing position information 32gnd of the photographed image 32gnb matches the position information of a specific location. If it is a specific shooting location, a type of costume corresponding to the shooting location is set, and even if it is not a specific shooting location, a type of costume corresponding to the shooting time is set. Then, a subject person image Hn to be synthesized (simulated) is extracted from the photographed image 32gnb, and the setting-type faceless costume image having a pose imitating the subject person image Hn is acquired, and the subject of the photographed image 32gnb is obtained. Since a composite image is created and distributed by replacing the person image Hn with the acquired faceless outfit image, the type corresponding to the shooting location and shooting time of the shot image 32gnb shot at each user terminal This costume image can be replaced and synthesized as a costume image with a pose imitating the subject person image Hn without being uncomfortable.

  In each of the above-described embodiments, the case has been described in which various composite images are generated and output using a photographed image obtained by a digital camera or the like as an original image. It is also possible to target an image already recorded on the other recording medium. Here, if each subject person image cannot be extracted in the target image extraction processing (method 1) (method 2) because there is no shooting information from the original image, contour enhancement processing is performed on the original image. Thus, it may be configured such that each subject person image is extracted by extracting a contour region including a wide skin color portion or extracting a contour region including a circular portion forming the human head.

  It should be noted that each processing method by the image output apparatus 10 (image distribution server 30) described in each of the embodiments, that is, the image synthesis process shown in the flowchart of FIG. 9 and the object associated with the image synthesis process shown in the flowchart of FIG. Each method such as an image extraction process, a corresponding costume image acquisition process associated with the image composition process shown in the flowchart of FIG. 11, and a similarity evaluation process with the imitation target person image associated with the corresponding costume image acquisition process shown in the flowchart of FIG. These are programs that can be executed by a computer, such as memory cards (ROM cards, RAM cards, etc.), magnetic disks (floppy disks, hard disks, etc.), optical disks (CD-ROM, DVD, etc.), external recording such as semiconductor memories, etc. It can be stored and distributed on the medium 13 (33). Various computer terminals having a communication function with the communication network (Internet) N transfer the program stored in the external recording medium 13 (33) to the memory 12 (32) by the recording medium reading unit 14 (34). By reading and controlling the operation by the read program, the image composition function described in the above embodiment can be realized, and the same processing by the above-described method can be executed.

  The program data for realizing each of the above methods can be transmitted on a communication network (Internet) N in the form of a program code, and the above-mentioned data can be transmitted from a computer terminal connected to the communication network (Internet) N. It is also possible to implement the above-described image composition function by fetching program data.

  Note that the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention at the stage of implementation. Further, each of the embodiments includes inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in each embodiment or some constituent features are combined, the problems described in the column of the problem to be solved by the invention can be solved. When the effects described in the column of the effect of the invention can be obtained, a configuration in which these constituent elements are deleted or combined can be extracted as an invention.

1 is a block diagram showing a configuration of an electronic circuit of an image output apparatus 10 according to a first embodiment of the present invention. Costume 1 “period drama” pose image memory 12 c of the image output device 10: Costume 2 “Western” pose image memory 12 d... Costume 13 “Summer (June-September)” The figure which shows the costume images 12cM, 12dM, and 12fM of each male version corresponding to a pose. Costume 1 “period drama” pose image memory 12 c of the image output device 10: Costume 2 “Western” pose image memory 12 d... Costume 13 “Summer (June-September)” The figure which shows the costume image 12cW, 12dW, and 12fW of each female version corresponding to a pose. The figure which shows the model image corresponding to the model pose memorize | stored in the model pose image memory 12b of the said image output apparatus 10. FIG. The figure which shows the replacement | exchange (dressing) synthetic | combination state of the costume pose image according to the imaging | photography place (period play village) with respect to the synthetic | combination (simulation) object person image of the said image output device 10. FIG. The figure which shows the replacement | exchange (dressing) synthetic | combination state of the costume pose image according to the imaging | photography place (Western village) with respect to the synthetic | combination (simulation) object person image of the said image output device 10. FIG. The figure which shows the replacement | exchange (dressing) synthetic | combination state of the costume pose image according to the imaging | photography season (year-month-day time) with respect to the synthetic | combination (simulation) object person image of the said image output device 10. FIG. FIG. 2 is a diagram showing an external configuration of a digital camera 10 that is the image output device and its cradle 15x. 4 is a flowchart showing image composition processing of the image output apparatus 10; 4 is a flowchart showing a target image extraction process for extracting a person image to be synthesized (simulated) in accordance with the image synthesis process of the image output apparatus 10, and FIG. 4A is a flowchart showing the target image extraction process (method 1). FIG. 5B is a flowchart showing the target image extraction process (method 2). It is a flowchart which shows the corresponding costume image acquisition process accompanying the image composition process of the said image output apparatus 10, The figure (A) is a flowchart which shows the corresponding costume image acquisition process (method 1), The figure (B) is the correspondence. The flowchart which shows a costume image acquisition process (scheme 2). The flowchart which shows the similarity evaluation process with the imitation object person image in the corresponding costume image acquisition process (method 1) (method 2) accompanying the image composition processing of the image output apparatus 10. The figure which shows the example of similarity evaluation with the costume pose image of the setting kind accompanying the similarity evaluation process in the corresponding | compatible costume image acquisition process (method 1) of the said image output device. The figure which shows the similarity evaluation example with the model image accompanying the similarity evaluation process in the corresponding | compatible clothing image acquisition process (method 2) of the said image output device. The block diagram which shows the structure of the electronic circuit of the image delivery server 30 which concerns on 2nd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Image output apparatus 10A ... Cell-phone with a digital camera 10B ... Digital camera with a communication function 30 ... Image delivery server 11, 31 ... CPU
12, 32 ... Memory 12a, 32a ... Image output processing program 12b, 32b ... Model pose image memory 12c, 32c ... Costume 1 "period drama" pose image memory 12d, 32d ... Costume 2 "Western" pose image memory 12e, 32e ... Costume 3 "Medieval Europe" pose image memory 12f ... Costume 13 "Summer (June-September)" pose image memory 12g ... Costume 14 "Spring autumn (4, 5, 10 and November)" pose image memory 12h ... Costume 15 “Winter (December-March)” pause image memory 32 gn ... Shooting image file 32 gna ... User name: File name: User authentication data 12i, 32 gnb ... Shooting image memory 12j, 32 gnc ... Compositing target image position memory 12k ... Position information registration memory 32gnd ... Shooting position information memory 12l, 32f ... Current date memory 12 m ... image processing work area 13, 33 ... external recording medium 14, 34 ... recording medium reading unit 15, 35 ... transmission control unit 16 ... input unit 17 ... image input unit 17A ... digital camera 17B ... digital video movie 18 ... distance sensor 18x: Position information detection unit 19, 36 ... Display unit 20 ... Image output unit 20A ... Memory card 20B ... VTR (recording device)
20C: Image printer 21: Web server N: Communication network (Internet)

Claims (15)

A captured image acquisition means for acquiring a captured image;
Target image extraction means for extracting a person image to be synthesized from the captured image acquired by the captured image acquisition means;
Position information obtaining means for obtaining position information obtained by photographing the photographed image;
Pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses in a type of costume according to position information;
A costume image storing a pose corresponding to the photographic position information of the photographic image acquired by the position information acquisition unit and corresponding to the person image to be synthesized extracted by the target image extraction unit. Corresponding image acquisition means for acquiring from among a series of costume images stored by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extracting means, leaving the face portion of the person image. ,
Image output means for outputting a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized by the image synthesizing means;
An image output apparatus comprising: A captured image acquisition means for acquiring a captured image;
Target image extraction means for extracting a person image to be synthesized from the captured image acquired by the captured image acquisition means;
Time information acquisition means for acquiring time information of photographing the photographed image;
Pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses with different types of costumes according to time information;
A costume image for each pose corresponding to the photographic time information of the photographic image acquired by the time information acquisition means, and a pose costume image corresponding to the person image to be synthesized extracted by the target image extraction means. Corresponding image acquisition means for acquiring from among a series of costume images stored by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extracting means, leaving the face portion of the person image. ,
Image output means for outputting a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized by the image synthesizing means;
An image output apparatus comprising:   3. The target image extracting unit extracts a changed portion as a human image to be synthesized from a series of a plurality of captured images acquired by the captured image acquiring unit. Image output device.   3. The target image extraction unit extracts a human image to be synthesized according to distance information between image portions of the captured image stored by the captured image acquisition unit. The image output device described. The corresponding image acquisition means includes
A similarity evaluation means for evaluating the similarity between the person image to be synthesized in the photographed image extracted by the target image extraction means and the costume image for each pose stored by the pose-specific costume image storage means;
The costume image of the pose corresponding to the person image to be synthesized is acquired according to the similarity between the person image to be synthesized and each costume image evaluated by the similarity evaluation means. The image output device according to claim 4. The corresponding image acquisition means includes
Model image storage means for storing model images of a plurality of poses associated with each pose of a series of costume images stored by the pose-specific costume image storage means;
Similarity evaluation means for evaluating the similarity between the person image to be synthesized in the captured image extracted by the target image extraction means and the model image of each pose stored by the model image storage means;
Acquire a costume image stored in the pose-specific costume image storage means in association with the pose of the model image in accordance with the similarity between each model image and the person image to be synthesized evaluated by the similarity evaluation means. The image output apparatus according to claim 1, wherein the image output apparatus is an image output apparatus. Equipped with shooting means,
The captured image acquisition means acquires a captured image captured by the imaging means;
The image output apparatus according to claim 1, wherein the image output apparatus is an image output apparatus. A captured image acquisition step of acquiring a captured image;
A target image extraction step of extracting a person image to be synthesized from the captured image acquired in the captured image acquisition step;
A position information acquisition step of acquiring position information obtained by capturing the captured image;
A pose-specific costume image storage step for storing a series of costume images consisting of a plurality of poses in a type of costume according to position information in a memory;
The costume image of the pose corresponding to the shooting position information of the shot image acquired in the position information acquisition step and corresponding to the person image to be synthesized extracted in the target image extraction step A corresponding image acquisition step of acquiring from a series of costume images stored in the memory in the image storage step;
Image composition for combining the costume image acquired in the corresponding image acquisition step with the person image to be combined in the photographed image extracted in the target image extraction step, leaving the face portion of the person image. Steps,
An image output step of outputting a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized in this image synthesis step;
An image output method comprising: A captured image acquisition step of acquiring a captured image;
A target image extraction step of extracting a person image to be synthesized from the captured image acquired in the captured image acquisition step;
A time information acquisition step of acquiring time information of capturing the captured image;
A pose-specific costume image storage step for storing a series of costume images of a plurality of poses in a type of costume according to time information in a memory;
The costume image corresponding to the shooting time information of the captured image acquired in the time information acquisition step and the pose according to the person image to be combined extracted in the target image extraction step A corresponding image acquisition step of acquiring from a series of costume images stored in the memory in the image storage step;
Image composition for combining the costume image acquired in the corresponding image acquisition step with the person image to be combined in the photographed image extracted in the target image extraction step, leaving the face portion of the person image. Steps,
An image output step of outputting a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized in this image synthesis step;
An image output method comprising: An image output processing program for controlling a computer to perform image output processing,
The computer,
Captured image acquisition means for acquiring captured images,
Target image extraction means for extracting a person image to be synthesized from the captured image acquired by the captured image acquisition means;
Position information acquisition means for acquiring position information of the captured image;
Pose-specific costume image storage means for storing a series of costume images of a plurality of poses in a type of costume according to position information in a memory;
A costume image storing a pose corresponding to the photographic position information of the photographic image acquired by the position information acquisition unit and corresponding to the person image to be synthesized extracted by the target image extraction unit. Corresponding image acquisition means for acquiring from a series of costume images stored in the memory by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extraction means, leaving the face portion of the person image;
Image output means for outputting a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized by the image synthesizing means;
A computer-readable image output processing program designed to function as a computer. An image output processing program for controlling a computer to perform image output processing,
The computer,
Captured image acquisition means for acquiring captured images,
Target image extraction means for extracting a person image to be synthesized from the captured image acquired by the captured image acquisition means;
Time information acquisition means for acquiring time information of photographing the photographed image;
Pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses in a type of costume according to time information in a memory;
A costume image for each pose corresponding to the photographic time information of the photographic image acquired by the time information acquisition means, and a pose costume image corresponding to the person image to be synthesized extracted by the target image extraction means. Corresponding image acquisition means for acquiring from a series of costume images stored in the memory by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extraction means, leaving the face portion of the person image;
Image output means for outputting a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized by the image synthesizing means;
A computer-readable image output processing program designed to function as a computer. A captured image receiving means for receiving a captured image from the terminal;
Target image extracting means for extracting a person image to be synthesized from the photographed image received by the photographed image receiving means;
Position information receiving means for receiving position information obtained by photographing the photographed image from the terminal;
Pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses in a type of costume according to position information;
The pose-specific costume image memory stores a pose costume image corresponding to the photographic position information of the photographic image received by the position information receiving means and corresponding to the person image to be synthesized extracted by the target image extracting means. Corresponding image acquisition means for acquiring from among a series of costume images stored by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extracting means, leaving the face portion of the person image. ,
A synthesized image transmitting means for transmitting to the terminal a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized by the image synthesizing means;
An image delivery server comprising: A captured image receiving means for receiving a captured image from the terminal;
Target image extracting means for extracting a person image to be synthesized from the photographed image received by the photographed image receiving means;
Time information receiving means for receiving time information of photographing the photographed image from the terminal;
Pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses with different types of costumes according to time information;
The pose-specific costume image memory stores a pose costume image corresponding to the photographic time information of the photographic image received by the time information receiving means and corresponding to the person image to be synthesized extracted by the target image extracting means. Corresponding image acquisition means for acquiring from among a series of costume images stored by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extracting means, leaving the face portion of the person image. ,
A synthesized image transmitting means for transmitting to the terminal a synthesized image in which a costume image of a pose corresponding to the person image to be synthesized is synthesized by the image synthesizing means;
An image delivery server comprising: An image distribution processing program for controlling a computer to perform image distribution processing,
The computer,
Captured image receiving means for receiving captured images from the terminal,
Target image extraction means for extracting a person image to be synthesized from the photographed image received by the photographed image receiving means;
Position information receiving means for receiving position information obtained by photographing the photographed image from the terminal;
Pose-specific costume image storage means for storing a series of costume images of a plurality of poses in a type of costume according to position information in a memory;
The pose-specific costume image memory stores a pose costume image corresponding to the photographic position information of the photographic image received by the position information receiving means and corresponding to the person image to be synthesized extracted by the target image extracting means. Corresponding image acquisition means for acquiring from a series of costume images stored in the memory by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extraction means, leaving the face portion of the person image;
A composite image transmitting means for transmitting, to the terminal, a composite image in which a costume image of a pose corresponding to the person image to be combined is combined by the image combining means;
A computer-readable image distribution processing program that functions as a computer. An image distribution processing program for controlling a computer to perform image distribution processing,
The computer,
Captured image receiving means for receiving captured images from the terminal,
Target image extraction means for extracting a person image to be synthesized from the photographed image received by the photographed image receiving means;
Time information receiving means for receiving time information of photographing the photographed image from the terminal;
Pose-specific costume image storage means for storing a series of costume images consisting of a plurality of poses in a type of costume according to time information in a memory;
The pose-specific costume image memory stores a pose costume image corresponding to the photographic time information of the photographic image received by the time information receiving means and corresponding to the person image to be synthesized extracted by the target image extracting means. Corresponding image acquisition means for acquiring from a series of costume images stored in the memory by the means;
Image combining means for combining the costume image acquired by the corresponding image acquisition means with the person image to be combined in the photographed image extracted by the target image extraction means, leaving the face portion of the person image;
A composite image transmitting means for transmitting, to the terminal, a composite image in which a costume image of a pose corresponding to the person image to be combined is combined by the image combining means;
A computer-readable image distribution processing program that functions as a computer.

Images