JP2007249371A - Method, program, and device for reproducing slide show, and image composition method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To display persons in a plurality of image data together for enhancing rendering effect of a slide show and increasing an added value of the slide show. <P>SOLUTION: Face images are extracted from image data reproduced in the slide show, and the extracted face images are grouped in the descending order of similarity. The number of face images in each group is counted. A face image with the largest size among the face images of each group is selected as a representative image, and in generation of an assembly scene, the sizes of the representative images are decided by ranking the representative images according to the number of face images in their groups. The assembly scene is inserted in the last scene of the slide show to be reproduced. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a slide show playback method, a slide show playback program, a slide show playback apparatus, and an image composition method for combining a plurality of image data in which a person is photographed.

  There is known a slide show in which a plurality of photographs taken with a digital camera are reproduced and displayed on a monitor while switching at predetermined intervals (see, for example, Patent Document 1). The slide show is suitable for viewing a large number of collected photos at once because a plurality of photos are automatically switched and displayed.

In the slide show playback method disclosed in Patent Document 1, not only a plurality of photos are simply switched, but also a zooming process that closes up a part of one photo, and panning that moves the viewpoint from one point of the photo to another point. A special effect such as processing is applied to produce a slide show. In recent years, a face detection technique for detecting a face image of a person shown in a photograph has been developed (see, for example, Patent Document 2). In the slide show reproduction method of Patent Document 1, such a face detection technique and the special effect are combined to produce effects such as close-up of a face image or panning from one face image to another.
JP 2005-182196 A JP-A-2005-174308

  As described above, the slide show reproduction technology is progressing in the direction of further increasing the added value of the slide show, and an idea for increasing the added value has been demanded in the development site.

  An object of the present invention is to increase the added value of a slide show.

  In order to achieve the above object, a slide show playback method of the present invention is a slide show playback method for playing back a slide show of a plurality of image data in which a person is photographed, and reads a plurality of image data from an image storage medium. A step, a face image extraction step for extracting a human face image from each read image data, a composite image generation step for generating a composite image in which a plurality of extracted face images are arranged, and the composite image is a set of face images And a collective scene insertion step of inserting the entire slide show as a scene.

  Note that the collective scene is preferably inserted at a predetermined position in the entire slide show.

  In particular, it is preferably inserted as the last scene of a slide show.

  In addition, after the face image extraction step, a face image classification step for classifying the plurality of extracted face images into a plurality of groups according to the degree of similarity and a representative image used for the composite image are selected one by one from each group. A representative image selection step, and in the composite image generation step, it is preferable to generate a composite image by combining the plurality of selected representative images.

  Further, each representative image is preferably a face image having the largest size among the face images in each group.

  Alternatively, each representative image is preferably a face image classified first among the face images in each group.

  And a face image counting step for counting the number of face images in each group after the face image classification step. In the composite image generation step, the representative image of each group is displayed according to the number of face images counted. It is preferable to assign an order and determine the display size or layout of each representative image in the composite image according to the order.

  Furthermore, the composite image generation step is preferably performed before the slide show reproduction is started.

  A slide show reproduction program of the present invention is a slide show reproduction program for causing a computer to execute a slide show of a plurality of image data in which a person is photographed, an image data reading step for reading a plurality of image data from an image storage medium, and each read A face image extraction step for extracting a person's face image from image data, a composite image generation step for generating a composite image in which a plurality of extracted face images are arranged, and a combined scene of the person during the entire slide show And a collective scene insertion step of inserting into the slide show.

  A slide show playback apparatus according to the present invention is a slide show playback apparatus for playing back a slide show of a plurality of image data in which a person is photographed, an image data reading means for reading a plurality of image data from an image storage medium, and each read image data A face image extracting means for extracting a person's face image from the image, a composite image generating means for generating a composite image in which a plurality of extracted face images are arranged, and a slide show as a set scene of the person during the entire slide show And a set scene insertion means for inserting into the scene.

  The image composition method of the present invention is an image composition method for compositing a plurality of image data in which a person is photographed, an image data reading step for reading a plurality of image data from an image storage medium, and a person from each read image data A face image extracting step for extracting the face image, and a composite image generating step for generating a composite image in which the plurality of extracted face images are arranged.

  According to the slide show reproduction method of the present invention, an image data reading step for reading a plurality of image data from an image storage medium, a face image extracting step for extracting a human face image from each read image data, and a plurality of extracted plural images A composite image generation step for generating a composite image in which face images are arranged, and a collective scene insertion step for inserting the composite image as a collective scene of the face image into the entire slide show, thereby enabling a plurality of persons of image data to be obtained. They can be displayed together, and the effect of the slide show effect is increased.

  In addition, since the collective scene is inserted at a predetermined position in the entire slide show, a certain effect can be obtained when reproducing any slide show.

  Further, after the face image extraction step, a face image classification step for classifying the plurality of extracted face images into a plurality of groups according to the degree of similarity and a representative image to be used for the composite image are selected one by one from each group. And a representative image selection step. In the composite image generation step, a plurality of selected representative images are combined to generate a composite image, so that the same person does not appear repeatedly in the collective scene.

  Furthermore, since each representative image is a face image having the largest size among the face images in each group, a face image having a good overall appearance is used, and a high-quality slide show can be reproduced.

  Alternatively, each representative image is a face image that is first classified among the face images in each group, so that it is processed uniformly, leading to a reduction in processing time.

  And a face image counting step for counting the number of face images in each group after the face image classification step. In the composite image generation step, the representative image of each group is displayed according to the number of face images counted. By adding an order and determining the display size or layout of each representative image in the composite image according to the order, the appearance frequency of each person in the slide show can be easily determined.

  A slide show playback apparatus 11 shown in FIG. 1 is an apparatus that plays back a slide show by switching a plurality of images displayed on a monitor 15, and is realized by installing a slide show playback program in a personal computer, for example. The slide show playback device 11 includes a CPU 12, a card reader 13, a RAM 14, a monitor 15, an operation unit 16, and a data storage device (hereinafter simply referred to as “storage”) 17. The CPU 12 comprehensively controls the entire apparatus according to operation instructions input from the operation unit 16. The storage 17 stores a slide show reproduction program (hereinafter simply referred to as “program”) 10. The CPU 12 loads the program 10 onto the RAM 14 and executes the processing steps described in the program 10, thereby reproducing the slide show. The RAM 14 is a working memory that is used by the CPU 12 to execute processing.

  The monitor 15 displays a slide show and various operation screens of the program 10. The operation screen includes an image selection screen for selecting images to be used for the slide show, a template selection screen described later, and the like. The memory card 18 is an image storage medium that stores images taken by a digital camera as digital data. A card reader (image data reading means) 13 accesses the memory card 18 and reads an image.

  When the program 10 is activated and an image to be used for the slide show is selected by the user on the image selection screen and a reproduction instruction is given, the card reader 13 reads the selected images Pic1 to PicX. When the slide show is started, the plurality of read images Pic1 to PicX are sequentially displayed on the monitor 15 while being switched at predetermined time intervals.

  The template is decoration data prepared in advance in order to give a special effect such as masking processing to the image data to produce a slide show. For example, as shown in FIG. 2, the masking process is a process in which a mask is put on the screen and the mask is partially removed. For example, as shown in FIG. 2 (d), when masking processing is performed on an image Pic100 in which three persons A, B, and C are shown, the entire screen is masked. As shown in FIG. 2A, the mask of the portion of the person A is removed, and then the mask is removed in the order of the person B and the person C as shown in FIGS. 2B and 2C. . By applying the mask, it is possible to make the portion where the mask is removed relatively prominent.

  There are other variations of the masking process, such as a process of moving the display frame that is out of the mask within the screen. FIG. 3 shows an example in which this process is applied to an image Pic200 in which the faces of three persons A, B, and C are arranged on a substantially straight line. As shown in FIG. 3 (a), first, the leftmost person A is displayed in the display frame, and then the display frame moves in the horizontal direction to display the middle person C as shown in FIG. 3 (b). Further, the display frame moves to the right, and the rightmost person B is displayed as shown in FIG.

  Various types of template data are prepared for each type of special effect, and are stored in the storage 17 together with the program 10. The user can select a template to be used for the slide show from these various templates from the template selection screen.

  In addition to the production of the slide show based on such a template, the slide show playback device 11 of the present invention has a collective scene insertion function for creating a collective scene of people appearing in the slide show and inserting it into the entire slide show. The collective scene is a composite image in which face images are extracted from a plurality of images used in a slide show and the extracted face images are arranged. By inserting such a collective scene into the slide show, it is possible to list the characters in the slide show.

  Since multiple images used in a slide show often include the same person across multiple images, simply extract the facial images of all the people from each image and synthesize those facial images. Alone, it is conceivable that face images of the same person overlap as in the collective scene 19 shown in FIG. Therefore, the slide show playback apparatus 11 of the present invention is devised to eliminate such duplication.

  Hereinafter, a case where a plurality of images Pic1 to PicX in which persons A to F appear is used for a slide show will be described as an example. As shown in FIG. 5A, the CPU (face image extraction means) 12 performs image analysis on each of the images Pic1 to PicX, and extracts the face images 21a to 21t included in the images Pic1 to PicX. . Extraction of the face image is performed by a method described in Patent Document 2, for example. Then, as shown in FIG. 5A and FIG. 5B, the extracted face images 21a to 21t are classified into groups 22a to 22f of persons A to F according to those having high similarity. Specifically, first, the face image 21a of the person A extracted from the image Pic1 is classified into a group 22a of the person A. Then, the face image 21b of the person B extracted from the image Pic2 is classified into the group 22b of the person B. Further, the face image 21c of the person C extracted from the image Pic3 is classified into the group 22c of the person C, and the face image 21d of the person A is classified into the group 22a of the person A. Hereinafter, the face images 21e to 21t extracted from the images Pic4 to PicX are similarly classified.

  Further, as shown in FIG. 6, the CPU (composite image generation means) 12 counts the number of face images in the groups 22 a to 22 f of the classified persons A to F, respectively. Then, the face images 21a, 21b, 21i, 21r, 21p, and 21n having the largest size among the face images in each group 22a to 22f are respectively selected as representative images. FIG. 7 shows representative images 21a, 21b, 21i, 21r, 21p, and 21n arranged in order from the representative images of the group having the largest number of face images and determining their display sizes according to the order. The collective scene 20 shown is generated. As described above, since one face image is selected as a representative image from each group classified for each face image of the same person, a collective scene in which the face images of the same person overlap is not generated. In addition, since the face image having the largest size among the face images in each group is selected as the representative image, the face image having a good overall appearance is used, and a high-quality slide show can be reproduced.

  Here, determination of the display size of the representative image will be described in detail. As shown in FIG. 6, the face images in the group 22a of the person A are counted as six images 21a, 21d, 21q, 21k, 21h, and 21l. Similarly, two face images 21b and 21o in the group 22b of the person B, three face images 21i, 21c and 21m in the group 22c of the person C, and face images in the group 22d of the person D Are 21r, 21j, four face images of the person E in the group 22e, 21p, 21s, 21e, and 21f, and three face images of the person F in the group 22f are 21n, 21t, and 21g. , Each is counted. Accordingly, the representative image 21p of the person E, the representative image 21i of the person C, the representative image 21n of the person F, the representative image 21b of the person B, in order from the representative image 21a of the person A having the largest number of face images in the group. The representative image 21r of the person D is ranked. Then, the representative image 21a of the person A is set to the largest display size, and the representative images 21p, 21i, 21n, 21b, and 21r are arranged as the display sizes are gradually reduced according to the order, thereby generating the collective scene 20 shown in FIG. Thereby, the appearance frequency of each person in the slide show can be easily determined.

  The operation of the slide show playback apparatus 11 having the above configuration will be described below with reference to the flowcharts of FIGS. When a slide show is played back by the slide show playback device 11, first, the memory card 18 is set in the card reader 13, and the images Pic 1 to PicX stored therein are read into the slide show playback device 11. The read images Pic1 to PicX are displayed on the monitor 15 as an image selection screen. The user operates the operation unit 16 while viewing the image selection screen, and selects an image to be reproduced in the slide show. Subsequently, a template selection screen for selecting a special effect template to be used for the slide show is displayed on the monitor 15. The user operates the operation unit 16 while viewing the template selection screen to select a template to be used for the slide show.

  The CPU 12 receives an operation instruction from the operation unit 16 and applies the special effect recorded in the template to the plurality of images Pic1 to PicX (for example, FIGS. 2 and 3).

  Further, the CPU 12 performs image analysis on each of the selected plurality of images Pic1 to PicX, extracts face images 21a to 21t included in the images Pic1 to PicX (FIG. 5A), and is extracted. The face images 21a to 21t are classified into groups 22a to 22f of persons A to F for each of those having a high degree of similarity (FIG. 5B).

  Furthermore, the CPU 12 counts the number of face images in the groups 22a to 22f of the persons A to F, respectively (FIG. 6). Then, the face images 21a, 21b, 21i, 21r, 21p, and 21n having the largest size among the face images in the groups 22a to 22f are respectively selected as representative images, and the representative of the group 22a having the largest number of face images is selected. The representative images 21p, 21i, 21n, 21b, and 21r are arranged in order from the image 21a. Further, the representative image 21a of the person A is set to the largest display size, and the representative images 21p, 21i, 21n, 21b, and 21r are arranged as the display sizes are gradually reduced according to the order, thereby generating the collective scene 20 (FIG. 7).

  When a reproduction instruction is given by the user's operation of the operation unit 16, the collective scene 20 is inserted into the last scene of the slide show and reproduced (FIG. 10 (a)). Thereby, it is possible to list the characters in the slide show without duplication. The collective scene 20 may be inserted at the beginning of the slide show (FIG. 10B) or may be inserted at an arbitrary position by the user's operation of the operation unit 16 (FIG. 10C). ).

  In the above-described embodiment, the case where the slide show playback apparatus 11 is realized by installing the program 10 in a personal computer has been described as an example. However, the present invention is not limited to this. good.

  Moreover, in the said embodiment, although the slide show demonstrated as an example the case where several images Pic1-PicX were displayed sequentially, switching automatically at predetermined time intervals, it is not limited to this, For example, operation The plurality of images Pic1 to PicX may be displayed while being switched by the operation of the unit 16.

  In the above embodiment, the memory card is described as an example of the image storage medium. However, various image storage media such as a CD, a DVD, and a USB memory may be used instead of the memory card.

  In the above embodiment, the case where the images used for the slide show are selected by the operation unit 16 has been described as an example. However, the present invention is not limited to this. For example, image data stored in an image storage medium is stored. It may be automatically loaded and all loaded are selected.

  In the above embodiment, the case where the image used in the slide show selects an image read by the card reader 13 has been described as an example. However, the present invention is not limited to this, and the image is stored in the storage 17 in advance. Alternatively, the image may be selected.

  In the above embodiment, the case where the collective scene 20 is generated after applying the special effect to the image has been described as an example. However, the special effect may be applied to the image after the collective scene 20 is generated. . In this case, special effects may be applied not only to the images used for the slide show but also to the collective scene 20.

  In the above-described embodiment, the collective scene is described as an example in which the scene is generated in advance before the slide show is played. However, the playback process and the generation process may be performed in parallel while the slide show is played.

  Further, in the above embodiment, the case where the face image having the largest size among the face images of each group is used as the representative image has been described as an example. However, for example, when the face image classification process is executed, The face images classified first may be used as representative images. Further, the focus state of each face image may be checked, and the one that is in focus may be used.

  Further, in the above embodiment, the case where the display size of each representative image in the collective scene is changed by ranking the representative images of each group according to the count number has been described as an example. It may be changed. In this case, for example, as in the above embodiment, the representative image 21p of the person E, the representative image 21i of the person C, the representative image 21n of the person F, the representative image 21b of the person B, and the person D in order from the representative image 21a of the person A. If the representative image 21r is ordered, the representative image 21a of the person A is laid out in the center, and the representative images 21p, 21i, 21n, 21b, 21r are gradually laid out from the center according to the order. 11 is generated.

  Further, in the above-described embodiment, the case where a composite image used for a slide show is generated has been described as an example. An image composition method to be generated is also included, and the use of the composite image is not limited to a slide show. In addition to the collective scene of the slide show, this composite image is used as print data, for example. Such a print of a composite image is useful for listing people in a plurality of images.

It is a block diagram which shows the structure of the slide show reproduction | regeneration apparatus of this invention. It is a figure explaining the masking process which produces image data. It is a figure explaining the modification of the masking process which produces image data. It is a figure of the collective scene where the face image of the same person overlapped. It is a figure explaining extraction and classification of a face image. It is a figure explaining the count of a face image. It is a figure of the collective scene which eliminated duplication of the face picture of the same person. It is a flowchart which shows the process sequence until reproducing | regenerating a slide show. It is a flowchart which shows the production | generation procedure of an assembly scene. It is a figure which shows the insertion position of the collective scene in the whole slide show. It is a figure of the collective scene in another embodiment.

Explanation of symbols

10 slide show playback program 11 slide show playback device 12 CPU
13 Card reader 14 RAM
15 Monitor 16 Operation Unit 17 Data Storage Device 18 Memory Card 19, 20, 23 Group Scene 21a-21t Face Image 22a-22f Group Pic1, Pic2, Pic3-PicX, Pic100, Pic200 Image

Claims (11)

In a slide show playback method for playing back a slide show of a plurality of image data in which a person is photographed,
An image data reading step of reading the plurality of image data from an image storage medium;
A face image extraction step for extracting a human face image from each read image data;
A composite image generation step for generating a composite image in which a plurality of extracted face images are arranged;
A slide show reproduction method comprising: a collective scene insertion step of inserting the composite image as a collective scene of the face image into the entire slide show.   2. The slide show playback method according to claim 1, wherein the collective scene is inserted at a predetermined position in the entire slide show.   3. The slide show reproduction method according to claim 2, wherein the collective scene is inserted as a last scene of the slide show. After the face image extraction step, a face image classification step for classifying the extracted plurality of face images into a plurality of groups according to the similarity, and one representative image to be used for the composite image is selected from each group one by one And a representative image selection step.
The slide show reproduction method according to claim 1, wherein, in the composite image generation step, the composite image is generated by combining the selected representative images.   5. The slide show reproduction method according to claim 4, wherein each representative image is a face image having a largest size among face images in each group.   5. The slide show playback method according to claim 4, wherein each of the representative images is a face image classified first among face images in each group.   A face image counting step for counting the number of face images in each group after the face image classification step, and in the composite image generating step, a representative image of each group is selected according to the number of face images counted. 7. The slide show reproduction method according to claim 4, wherein an order is assigned and a display size or layout of each representative image in the composite image is determined according to the order.   8. The slide show playback method according to claim 1, wherein the composite image generation step is performed before the slide show playback is started. In a slide show reproduction program for causing a computer to execute a slide show of a plurality of image data in which a person is photographed,
An image data reading step of reading the plurality of image data from an image storage medium;
A face image extraction step for extracting a human face image from each read image data;
A composite image generation step for generating a composite image in which a plurality of extracted face images are arranged;
A slide show reproduction program, comprising: a combined scene insertion step of inserting the composite image into the slide show as the collective scene of the person during the entire slide show. In a slide show playback device that plays back a slide show of a plurality of image data in which a person is photographed,
Image data reading means for reading the plurality of image data from an image storage medium;
A face image extracting means for extracting a person's face image from each read image data;
A composite image generation means for generating a composite image in which a plurality of extracted face images are arranged;
A slide show playback device comprising: a set scene insertion unit that inserts the composite image into the slide show as the set scene of the person during the entire slide show. In an image composition method for combining a plurality of image data in which a person is photographed,
An image data reading step of reading the plurality of image data from an image storage medium;
A face image extraction step for extracting a human face image from each read image data;
A composite image generation step of generating a composite image in which a plurality of extracted face images are arranged.

Images