JP2019079116A - Information processing device, information processing method, program and storage medium - Google Patents

Abstract

To maintain correlation between contents reproduced by a user and recommended contents.SOLUTION: An information processing device 1 has a recording device which records a plurality of event image groups which contain a plurality of images in which each image property information is associated with each other, and records recommendation reasons of recommended contents recommended to be presented to a user, after contents to be reproduced by the user are reproduced. The information processing device 1 further has: an event image group selection unit 205 which selects event image groups used for generating the recommended contents from among the event image groups, by using the recommendation reasons; and an image selection unit 206 which selects the images used for generating the recommended contents from among the images contained in the selected event image groups.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an information processing apparatus, an information processing method, and a program.

  In recent years, with the spread of digital cameras and smartphones, the number of images that can be photographed in digital cameras and the like is extremely large, and a content generation method using a large number of images is required. As an example of such a method, a folder (for example, "hot spring trip on March 19, 2016") with an event name related to the image is generated, and the image is classified and managed for each generated folder. The thing is mentioned.

  From a large number of images classified by the folder generated for each event in this manner, a slide show or a photo book can be given as content for browsing the image of a specific event. The user can view a large number of images by content such as a slide show or photo book.

  For example, a slide show is often reproduced in several seconds per image, and when the number of images is large, it may take a very long time to reproduce all the images in one event. Then, the method of patent document 1 is known as a method of narrowing down the number of images to reproduce | regenerate. In this method, each image included in a certain image group is analyzed to generate an evaluation score of the image, and an image is automatically selected based on the generated evaluation score.

  In addition, with the spread of digital cameras and smartphones, shooting has become easy, and not only the number of images shot but also the number of events tends to increase. In such an environment, methods are known for viewing and enjoying images of various events taken by the user.

  For example, there is known a method of automatically generating a slide show by specifying an event which is supposed to be desired by the user in relation to the slide show being played while the user is playing a slide show. In this approach, automatically generated slide shows can also be recommended to the user.

  When presenting the user with content that is supposed to be viewed by the user as recommended content, it is important to inform the user of the relevance between the content being played back and the recommended content. As a method of realizing this, a method using similarity between events is generally known. For example, Patent Document 2 selects moving image content of another event similar to the event based on event information of moving image content being reproduced, and recommends it to the user as moving image content having relevance. We disclose the method.

  Further, an event having a connection with the content (slide show) being reproduced is selected using the method of Patent Document 2, and a predetermined number of images are selected from images of the selected event using the method of Patent Document 1 You can also

  Also in the photo book, since the time for which the user views the image is limited, it is preferable to narrow the number of images to a number that allows the user to view from a large number of images. Therefore, as in the example of the slide show described above, it is preferable to recommend a photobook that is presumed to be browsed by the user due to the similarity between events and the like.

  In this way, even when there are a large number of captured images, automatically generate content such as a slide show or a photo book that is supposed to be viewed by the user, and present the generated content as recommended content to the user. Can.

JP, 2015-089112, A JP, 2011-217209, A

  The method of Patent Document 1 can narrow down the number of images, but does not disclose a method of generating content having relevance to the content to be reproduced by the user.

In the method of Patent Document 2, it is possible to select an event having relevance to the content specified based on the event information. However, when selecting an image from the image group belonging to the selected event, there is a risk that the relevance to the designated content may be lost depending on the selection method. An example will be described in which recommended content is generated by selecting an event having a relevance that characters are common to content reproduced by the user. In this case, if a character common to the photo included in the content reproduced by the user and the photo included in the recommended content is captured, the relationship that the characters are common is maintained.
However, for example, when the recommended content is generated by selecting an image such as a landscape image in which a common character is not shown because the image quality is emphasized, the relevance that the characters are common is lost. Therefore, it becomes difficult for the user to understand the relation between the content such as the slide show being played back and the recommended content.

  In addition, when the user browses images taken in large quantities, it is desirable to recommend content such as a slide show of various contents. For example, with respect to a slide show being played back, it is possible to recommend a slide show having relevance for the type of event and a slide show having relevance for characters. Therefore, it may be required to automatically generate and recommend a recommended slide show by increasing the types of event selection methods and image selection methods in order to realize contents of various contents. Even in this case, depending on the combination of event selection methods, there is a risk that the relevance between the content being played by the user and the recommended content may be lost.

  The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to maintain the relationship between content reproduced by a user and content recommended to be presented after reproduction of the content. .

  In order to solve the above problems, according to the present invention, an event image group recording unit in which a plurality of event image groups including a plurality of images each associated with image attribute information is recorded, and content to be reproduced by a user Event image used to generate the recommended content from the group of event images using the recommended reason recording means in which the recommended reason for recommended content recommended to be presented to the user after the recording is recorded, and the recommended reason Among the event image group selection means for selecting a group and the images included in the selected event image group, the image attribute information is referenced to obtain the image satisfying the recommendation reason, and the obtained image And image selection means for selecting the image used to generate the recommended content.

  According to the present invention, it is possible to maintain the association between content reproduced by a user and content recommended to be presented after reproduction of the content.

FIG. 2 is a hardware configuration diagram of an information processing apparatus. FIG. 2 is a functional block diagram of an information processing apparatus. The figure showing an event picture group table. The figure showing an image table. Diagram showing the recommended reason table. FIG. 8 is a diagram showing an event image group selection condition table. The figure showing an image condition table. The figure showing a slide show management table. The flowchart which shows the operation | movement which recommends a slide show. An explanatory view of a screen of a recommended slide show. FIG. 2 is a functional block diagram of an information processing apparatus. Diagram showing the recommended reason table. The figure showing a representative image selection condition table. 6 is a flowchart illustrating processing of displaying a representative image. Explanatory drawing of the screen of the slide show which displayed the representative image. FIG. 8 is a diagram showing an event image group selection condition table. The figure showing an image selection condition table. The flowchart showing the operation which recommends a slide show.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The configurations shown in the following embodiments are merely examples, and the present invention is not limited to the illustrated configurations. In the first to third embodiments described below, an example is shown in which the information processing apparatus according to the present invention is applied to a slide show recommendation apparatus that generates a recommended slide show for a user and presents or displays the slide show. The recommended slide show refers to a slide show that is recommended to be presented or displayed to the user after the content is played back, such as a slide show. However, the present invention is not limited to the slide show recommendation device. For example, the present invention can be applied not only to a slide show but also to an information processing apparatus that presents, to a user, arbitrary content related to an image such as a photo book or a photo movie as recommended content.

First Embodiment
FIG. 1 is a diagram showing a hardware configuration of the information processing apparatus 1 according to the first embodiment. The information processing apparatus 1 is a computer including a central processing unit (CPU) 101, a random access memory (RAM) 102, and a read only memory (ROM) 103. The CPU 101, the RAM 102, and the ROM 103 are connected to the input / output bus 111, and can mutually transmit and receive data. To the input / output bus 111, a network interface NETIF 104, a video RAM VRAM 105, an input / output controller 107, an HDD (Hard Disk Drive) 109, and an input / output IF 110 are connected. A display 106 which is a display device is connected to the VRAM 105, and an input device 108 is connected to the input / output controller 107.

  The CPU 101 is a central processing unit that controls the information processing apparatus 1. The CPU 101 executes a control program read from the ROM 103 to control the operation of the information processing apparatus 1 in an integrated manner, and executes calculation and processing of information and control of each hardware. Thus, the CPU 101 implements each functional configuration and processing described later.

  The RAM 102 is a random access memory, and functions as a main memory of the CPU 101 as a work memory necessary for loading an execution program and executing a program. The ROM 103 is a read only memory in which a control program that defines an operation processing procedure of the CPU 101 is recorded. The ROM 103 has a program ROM storing basic software (OS: Operating System) which is a system program for controlling the devices of the computer system, and a data ROM storing information necessary for operating the system. Note that OS data or the like may be recorded in the HDD 109 without using the ROM 103.

  The NETIF 104 is a network interface and performs input / output control of data such as image data transmitted / received via a network (not shown). The VRAM 105 is a video RAM in which image data representing an image displayed on the screen of the display 106 as a display device is expanded.

  The display 106 is a display device such as a liquid crystal TV, and is a liquid crystal display in the first embodiment. The input / output controller 107 is a controller that controls an input signal from the input device 108. The input device 108 is an arbitrary external input device for receiving an operation instruction from a user, and examples thereof include a touch panel, a keyboard, a mouse, and a remote control. The HDD 109 is a hard disk drive and a recording device. The HDD 109 stores necessary data such as an application program, image data of an image to be reproduced in a slide show, and information to be described later associated with each image.

  The input / output IF 110 is an interface for connecting to an external device such as a CD (Compact Disc) -ROM drive, a DVD (Digital Versatile Disc) drive, or a memory card drive. The input / output IF 110 is used, for example, for reading image data captured by a digital camera. The input / output bus 111 is a bus (such as an address bus, a data bus, and a control bus) for connecting the units described above.

  The information processing apparatus 1 generates a slide show list for specifying images to be reproduced in a slide show and the reproduction order thereof, and reproduces the slide show according to the slide show list.

  In this embodiment, the information processing apparatus 1 reproduces the slide show and displays it on the display 106. However, as another form, the slide show may be reproduced by a slide show reproduction device separate from the information processing device 1. In this case, the information processing device 1 transmits necessary data such as a slide show list and an image used for the slide show to the slide show reproduction device.

  FIG. 2 is a functional block diagram of the information processing apparatus 1 according to the first embodiment. The information processing apparatus 1 includes an event classification unit 201, an event attribute information management unit 202, an image attribute information management unit 203, a recommendation reason determination unit 204, an event image group selection unit 205, and an image selection unit 206 as its functional blocks. The information processing apparatus 1 further includes a recommendation unit 207, a rule management unit 208, and a slide show management unit 209.

  Hereinafter, an outline of processing performed by each functional block in the information processing device 1 will be described. The event classification unit 201 classifies image data into an event image group when image data is stored in the HDD 109 or at an arbitrary timing. The event image group is an image group obtained by classifying a plurality of image data stored in the HDD 109 for each event, and a plurality of event images are recorded in the HDD 109.

  To classify image data, time information such as shooting start date and time for each image data, an event type representing an event for which the image was captured, and a character for identifying a person or a number of persons appearing in the image It is necessary to associate various information such as information and file name. These pieces of information can be acquired and associated, for example, by acquiring them from metadata in image data of an image captured according to an existing standard such as EXIF. In addition, such information is acquired and associated by any other method such that the user inputs information to image data at the time of imaging or automatically acquires the image data of the imaged image by image recognition or the like. can do.

  Any method can be used as a method of classifying image data for each event. For example, there is a method of classifying by date based on shooting date of image, a method of sorting image data by folder stored by the user, and a method of sorting by distribution of shooting date of image by an algorithm such as K average method. The event attribute information management unit 202 acquires attribute information of the event image group classified by the event classification unit 201, and manages the attribute information in the event image group table 301 of FIG. In the first embodiment, the event image group table 301 is recorded in the HDD 109 so that the event attribute information management unit 202 can read and write. Therefore, the HDD 109 has a function as an event image group recording unit.

The image attribute information management unit 203 acquires attribute information of the stored image data at the timing when the image data is stored in the HDD 109 or at any timing, and adds the attribute information to the image table 401 in FIG. The recommendation reason determination unit 204 determines a recommendation reason indicating the reason why the slide show is to be recommended, using the recommendation reason table 501 of FIG. 5.
In the first embodiment, the recommendation reason table 501 is recorded in advance in the HDD 109 so as to be readable by the recommendation reason determination unit. Therefore, the HDD 109 has a function as a recommended reason recording unit. The recommendation reason table 501 may allow deletion, addition and modification of a specific recommendation reason through the input device 108 according to the preference of the user. As an example of the change, the condition of "character similar" is set as "four or more characters" or "three or more characters" shown in FIG.

  The rule management unit 208 manages event conditions using the event image group selection condition table 601 of FIG. 6 and the image selection condition table 701 of FIG. 7. The event image group selection condition table 601 and the image selection condition table 701 are both readably stored in the HDD 109. Therefore, the HDD 109 functions as a condition recording unit in which the event image group selection condition is recorded and a condition recording unit in which the image selection condition is recorded. In the first embodiment, the rule management unit 208 reads these from the HDD 109 as necessary, and transmits them to the event image group selection unit 205 or the image selection unit 206. The event image group selection condition shown in FIG. 6 is a condition for determining whether to select an event image group.

  The event image group selection unit 205 uses the recommendation reason determined by the recommendation reason determination unit 204 to generate an event image group used to generate a recommended slide show from among the event image groups shown in the event image group table 301 of FIG. Choose Therefore, in the first embodiment, the event image group selection unit 205 acquires an event image group selection condition from the rule management unit 208. Then, an event image group having attribute information that matches the acquired event image group selection condition is selected from the event image group table 301.

The image selection unit 206 refers to attribute information among images included in the selected event image group to obtain the image satisfying the recommendation reason, and is used to generate a recommended slide show from among the obtained images. The image is selected.
Therefore, in the first embodiment, the image selection unit 206 acquires the image selection condition managed by the rule management unit 208 based on the recommendation reason, and the image having image attribute information matching the image condition is displayed in the image table 401. To choose from.

  The recommendation unit 207 generates a slide show list designating the reproduction order of the selected image, generates a recommended slide show from the selected image and slide show list, and outputs the slide show to the display 106. The slide show management unit 209 records the slide show list generated by the recommendation unit 207 in the HDD 109. Specifically, the slide show management unit 209 adds the slide show list generated by the recommendation unit 207 to the slide show management table 801 shown in FIG. The slide show management table 801 may add not only information generated by the recommendation unit 207 but also information on a slide show manually generated by the user.

  Next, details of each table shown in FIGS. 3 to 8 will be shown. The event image group table 301 in FIG. 3 is a table for managing attribute information of the event image group, and includes an event image group ID, shooting start date and time, event type, and characters. In the first embodiment, the shooting start date and time, the event type, and the characters are attribute information of the event image group. The event image group ID represents an ID of an event image group.

  The shooting start date and time is the shooting date and time of the first captured image included in the event image group, and the character is information of a person recognized in any of the images included in the event image group. The event type indicates what kind of event the event is, and one of “daily”, “travel”, and “party” is set. As a method of identifying an event type, there is a method of identifying using machine learning based on attribute information as a feature amount of an image included in an event image group. In addition to the above, other types such as “wedding”, “entrance ceremony”, “graduation ceremony”, “exercise event”, “festival”, etc. may be used.

  When describing an event image group whose event image group ID is 1 (hereinafter referred to as event image group 1), the shooting start date and time is 10:37 on March 2, 2014, and is included in event image group 1 The shooting date and time of the captured image will be this time or later. Further, since the event type is “everyday”, all the event types of the images included in the event image group 1 are “everyday”. On the other hand, the characters are “A, B, C, D”, and one of the images in the event image group 1 includes the person A. The same applies to the persons B, C, and D, and one of the images in the event image group 1 includes the persons B, C, and D. However, even if the event type is everyday, a scene in which a person does not appear may be photographed, so an image in which none of the persons A, B, C, and D are captured is included in the event image group 1 There is also.

  An image table 401 in FIG. 4 is a table for managing attribute information on each image included in the event image group, and includes an image ID, a file name, an event image group ID, shooting date and time, a scene and a person. The I event image group ID is an ID of an event image group to which an image belongs as described in FIG. A scene represents a photographed scene of an image, and one of "person" and "landscape" is set. In addition to the above, other scenes such as "cooking", "flower", and "building" may be used. The person is information of the person recognized from the image, but the person can be specified by a method such as direct input by the user.

  To describe image data whose image ID is 1 (hereinafter simply referred to as image data 1), the file name is IMG_0001_.JPG, and the event ID is 1. The shooting date and time is 10:37 on March 2, 2014, and the scene is "person" and "A, C, D" is included as a person. Therefore, the image data 1 is image data included in the event image group 1 of FIG. 3, and in the image, persons A, C, and D among the characters A, B, C, and D of the event image group 1 It is being imaged.

  The image data 2702 is image data included in the event image group 93 whose characters are "A, B, C, D, L, M", but the scene is a landscape and the person is "none". It has become. On the other hand, the image data 2703 is also an image included in the event image group 93, and the scene is a "person", and "A, B, C, D" among the characters in the event image group 93 are imaged as the person. ing. The recommendation reason table 501 of FIG. 5 is a table for managing the recommendation reason, and is configured by the recommendation reason ID, the recommendation reason, and the image group condition during reproduction. The event image group condition during reproduction is information for determining which recommendation reason is to be determined as the recommendation reason, and indicates the condition of the event image group to which the slide show being reproduced belongs.

The recommendation reason table 501 of FIG. 5 is a table for managing the recommendation reason, and has attribute information such as a recommendation reason ID, a recommendation reason, and a currently-replaying event image group condition. The recommended reason ID (hereinafter simply referred to as recommended reason 1) having a recommendation reason ID of 1 will be described.
In recommendation reason 1, the recommendation reason is "character similar", and the event image group condition during reproduction is "four or more characters". Therefore, when recommendation reason 1 is selected, the user is presented with the recommendation reason of character similarity. In the recommendation reason 2, the recommendation reason is “date similar”, and the event image group condition during reproduction is “event type is everyday”. In recommendation reason 3, the recommendation reason is “event type similar”, and the event image group condition during reproduction is “other”.

  The recommended reason determining unit 204 selects and determines a recommended reason for a recommended slide show from the group of event images being reproduced. For example, when the event image group being reproduced satisfies the condition that "four or more characters", the recommendation reason is 1, and the recommendation reason is "character similarity". If the event image group being reproduced satisfies the condition that “event type is everyday”, the recommended reason is 2 and the recommended reason is “event type is everyday”.

  If the slide show being played back does not satisfy any of these conditions, it corresponds to "Other", so the recommendation reason is 3 and the recommendation reason is "Event type similar". Therefore, at least recommendation reason 3 is selected as the recommendation reason no matter what event image group is being reproduced. From this, it can be understood that at least one recommendation reason is selected from the slide show being played back. The recommendation reason determination unit 204 determines at least one of the recommendation reasons selected as necessary as the recommendation reason used for generating the recommendation slide show.

  If the slide show being reproduced is a slide show generated from the event image group shown in FIG. 3, the recommendation reason determining unit 204 determines the recommendation reason with reference to the event image group table 301 of FIG. In other cases, the recommendation reason determination unit 204 selects and decides the recommendation reason by obtaining characters, event information, and the like from attribute information associated with the image included in the slide show being reproduced.

  The event image group selection condition table 601 of FIG. 6 is a table for managing the selection conditions of the event image group, and has attribute information such as an event image group selection condition ID, a recommendation reason ID, and an event image group selection condition. The recommendation reason ID is an ID of a recommendation reason corresponding to the event image group selection condition ID. The event image group selection condition is a condition for selecting an event image group used to generate a recommended slide show.

  The event image group selection condition 1 will be described. The corresponding recommendation reason ID is 1, and the event image group selection condition is “more than the number of designated common characters”. The event image group selection condition 2 has a corresponding recommended reason of 2, and the event image group selection condition is “event image group having the same type of event”. In the event image group selection condition 3, the corresponding recommended reason is 3, and the event image group selection condition is "an event image group having a difference in time of the shooting start date within two weeks". Therefore, when the event image group selection condition 1 is selected, the event image group selection unit 205 displays the characters in the image group whose character is being reproduced among the event image groups 1 to 93 shown in FIG. And select an event image group common to four or more people.

  The image selection condition table 701 in FIG. 7 is a table for managing the image selection conditions, and has attribute information such as an image selection condition ID, a recommendation reason ID, and an image selection condition. The recommendation reason ID represents an ID of a recommendation reason corresponding to the image selection condition ID. The image selection condition is an image selection condition used to generate a recommended slide show.

  The image selection condition 1 corresponds to the recommendation reason 1 and is "select image data of an image showing a common character". The image selection condition 2 corresponds to the recommendation reason 2 and is “select the ratio between the landscape image and the person image according to the event type”. In detail, if the event type is travel, the image selection condition is such that the image is selected at a ratio of 50% for the landscape image and 50% for the person image. Also, if the event type is everyday, the image is selected such that the landscape image is 20% and the person image is 80%, and if the event type is party the scene is 0% and the image is 0 and the person 100%. The image selection condition 3 corresponds to the recommendation reason 3 and is "randomly select image data". The image selection unit 206 refers to the recommendation reason and the image selection condition table 701 to select an image.

  The slide show management table 801 is a table for managing the generated recommended slide show, and includes a list ID, an event image group ID, a selected image ID, and a reproduction date and time. The slide show management unit 209 receives the slide show list generated by the recommended unit, adds the slide show list to the slide show management table 801 of FIG. The event image group ID is an ID of an image group selected corresponding to the slide show ID by the event image group selection unit 205, and the image reproduced in the recommended slide show is selected from the event image group.

  The selected image ID is an ID of an image selected corresponding to the image selection condition ID by the image selection unit 206 among the images constituting the event image group selected as described above. In the example of list 1, the event image group ID is 67, the selected image ID is 1801, 1806, 1809... 1891 and the reproduction date is 2015/10/10 18:10.

  When the recommendation unit 207 reproduces a recommended slide show, an image 1801, an image 1806, an image 1809,..., An image 1891 which are images recorded in the selected image ID are reproduced. The playback date is time information recorded when the user plays the recommended slide show. Therefore, the value of attribute information is not set in a recommended slide show that has only been generated but has not yet been reproduced.

Next, processing performed by the information processing device 1 at the time of reproduction of a recommended slide show will be described. Note that, unless otherwise noted, the following processing is executed by the CPU 101 through each functional block shown in FIG.
FIG. 9 is a flowchart showing an operation of generating and recommending a recommended slide show to be reproduced next from the slide show being reproduced.
The recommended slide show is generated as follows from the event image group to which the image included in the slide show being played belongs and the attribute information associated with the image.

  The event attribute information management unit 202 acquires attribute information of an event image group to which the slide show that has been reproduced last or is being reproduced belongs (S901). To that end, the event attribute information management unit 202 acquires from the slide show management table 801 the event image group ID of the slide show list with the latest reproduction date and time.

  In the first embodiment, the reproduction date and time of the list 19 is the newest, and the ID of the corresponding event image group is 93. The CPU 101 acquires the corresponding event image group 93, and acquires the attribute information, that is, the shooting start date and time, the event type, and the characters from the event image group table 301.

Referring to FIG. 3, the shooting start date and time of the event image group 93 is 2015/12/31 10:14, the event type is travel, and the characters are six persons of A, B, C, D, L, and M.
Next, the recommended reason determining unit 204 determines a recommended reason for the recommended slide show (S902). The recommendation reason determination unit 204 randomly selects one from the recommendation reason in which the attribute information of the event image group shown in FIG. 3 obtained in S901 and the event image group condition during reproduction shown in the recommendation reason table 501 match. Determine one recommendation reason.

  In FIG. 5, the event image group condition during reproduction of recommendation reason 1 includes four or more characters, an event type of everyday, and others. On the other hand, since six characters in the event image group 93 and the event type are travel, the recommended reason 1 of four or more characters is a candidate for the recommended reason. However, since the event type of the event image group 93 is travel, which does not correspond to “daily” of recommendation reason 2, recommendation reason 2 is excluded from the recommendation reasons. Furthermore, since the recommendation reason 3 is “other (optional)”, the event image group 93 also meets this condition, and the recommendation reason 3 is also a candidate for the recommendation reason.

  As described above, since there are a plurality of candidates for the reason for recommendation, in the first embodiment, any one of them is randomly determined as the reason for recommendation. Note that the method of determining the recommendation reason is not limited to the method of determining from the information of the event image group condition, and may simply be selected randomly, or even if the recommendation reason is selected according to the description order of the recommendation reason table 501 Good. In this example, recommendation reason 1 was determined as the recommendation reason.

  Thereafter, the event image group selection unit 205 acquires, from the event image group selection condition table 601, an event image group selection condition corresponding to the determined recommendation reason ID. In this example, an event image group selection condition corresponding to recommendation reason 1, that is, a condition that an event image group having a predetermined number or more of common characters is selected is acquired (S903).

  Next, the event image group selection unit 205 selects an event image group having event attribute information matching the acquired event image group selection condition from the event image group table 301 (S904). In this example, an event image group in which the number of common characters is a predetermined number or more is selected. Also, the predetermined number was 4 or more.

  When a plurality of event image groups can be selected, one may be selected at random, or ones whose shooting start date and time are the farthest from those being reproduced may be selected. In addition, an event image group having many matching attribute information may be preferentially selected. In this example, in the event attribute information table of FIG. 3, four or more characters in common with the event image group 93 are image data groups having event image group IDs of 1, 4 and 22.

  Among them, only the image data group of the event image group 4 has the same event type as the event image group 93 (travel together), so in this example, a recommended slide show is generated using the event image group 4 It was decided. The image selection unit 206 acquires an image selection condition corresponding to the ID of the determined recommendation reason from the image selection condition table 701 (S905). In this example, as the image selection condition corresponding to recommendation reason 1, an image selection condition of selecting an image in which a common character is shown is acquired.

The image selection unit 206 selects, from the image table 401, a predetermined number of images having the ID of the event image group selected in S904 and having the image attribute information matching the acquired image selection condition (S906).
The number to be selected may be calculated from the reproduction time of the slide show, or may be calculated from the number of images included in the event image group. If the number of images matching the conditions is larger than the number desired to be selected, the images may be narrowed randomly, or the image quality may be evaluated and images may be narrowed to high scores. Further, when the number of images matching the condition is smaller than the number desired to be selected, only the selected image may be used, or the image belonging to the event image group may be complemented. Alternatively, in this case, the process may return to the process of S903 to select another event image group.

  The recommendation unit 207 generates a recommended slide show using the selected image and slide show list (S 907). The recommended slide show is output to the display 106 so that playback is started when the slide show being played back is finished. In the first embodiment, the recommendation unit 207 outputs the recommendation reason used for generating the recommendation slide show, and displays the recommendation reason as a text to the user when the reproduction of the recommendation slide show starts. This makes it possible to reveal to the user the recommended reason for the recommended slide show.

  FIG. 10 is an explanatory diagram of a recommended slide show. The screen 1001 shows the last image of the slide show being played back, and the screen 1002 shows the first image of the slide show generated in S907. The recommendation unit 207 reproduces the last image on the screen 1001 and then reproduces the recommended slide show after displaying the screen 1002.

  In the first embodiment, since the attribute information of the event image group to which the slide show being reproduced belongs is used as the event image group selection condition and the image selection condition, the event attribute information is acquired in S901. As each selection condition, attribute information of an image included in the slide show being reproduced may be used, or the image attribute information may be acquired in S901.

  In each selection condition, the history of the slide show reproduced in the past may be acquired from the event ID and the image ID of the slide show management table 801, and the recently browsed event image group and image may not be selected. Furthermore, a new recommendation reason may be added to the recommendation reason table 501. In that case, each selection condition corresponding to the recommendation reason is added to the event image group selection condition table 601 and the image selection condition table 701.

  As described above, according to the first embodiment, it is possible to recommend various slide shows having a connection with the slide show being played based on various recommendation reasons, and the user can take an image taken without getting tired. It can be viewed.

Second Embodiment
In the first embodiment of the present invention, after selecting a single recommendation reason, one slide show is generated based on the selected recommendation reason and displayed to the user. However, the user may not necessarily be satisfied with the recommendation reasons selected in this way, and may desire a slide show generated based on other recommendation reasons.

  Therefore, in order to cope with this problem, in the second embodiment, a plurality of recommended reasons are selected, and representative images of the slide shows are displayed for the respective recommended reasons so that the user can select a desired slide show. Hereinafter, differences between the first embodiment and the second embodiment will be described.

  FIG. 11 shows a functional block diagram of the information processing device 3 according to the second embodiment. The information processing device 3 is obtained by adding a representative image selection unit 1101 for selecting a representative image of a slide show to the functional block diagram of the information processing device 1 shown in FIG. The operations of the event classification unit 201 to the recommendation unit 207 and the slide show management unit 209 except for the recommendation reason determination unit 204 and the like described below are the same as the operations with reference to FIG.

  The recommendation reason determination unit 204 determines a plurality of recommendation reasons using the recommendation reason table 1201 of FIG. 12 instead of the recommendation reason table 501 of FIG. 5. The rule management unit 208 manages representative image selection conditions in the representative image selection condition table 1301 of FIG. 13 in addition to the event image group selection conditions and the image selection conditions.

  The representative image selection unit 1101 acquires representative image selection conditions managed by the rule management unit 208 for each of the determined plurality of recommendation reasons. Then, for each recommendation reason, the representative image selection unit 1101 selects an image having image attribute information that matches the representative image selection condition as a representative image. The recommendation unit 207 displays the representative image selected for each recommendation reason and the recommendation reason together to the user, and reproduces the slide show corresponding to the representative image selected by the user through the input device 108.

  The recommendation reason table 1201 shown in FIG. 12 is a table for managing the recommendation reason, and includes the ID, the recommendation reason, and the display position in the second embodiment. The display position is information indicating which display position on the screen on which a slide show is recommended to the user is displayed and which recommendation reason.

  When the recommendation reason 1 is described as an example, the recommendation reason is "character similar" and the display position is "left". In the recommendation reason 2, the recommendation reason is "event type similar", the display position is "center", and in the recommendation reason 3, the recommendation reason is "date similarity", and the display position is "right". The event image group condition during reproduction is the same as the recommended reason table 501 in FIG. The recommendation reason table 1201 is used when determining the recommendation reason in the operation flow of FIG. 14 described later.

  In the second embodiment, a representative image selection condition table 1301 shown in FIG. 13 is newly used. The representative image selection condition table 1301 is a table for managing selection conditions of representative images, and includes a representative image selection ID, a recommendation reason ID, and a representative image selection condition. The recommendation reason ID is an ID of a recommendation reason corresponding to the representative image selection condition.

  The representative image selection condition is a selection condition of a representative image of the slide show. The representative image selection condition 1 corresponds to the recommendation reason 1, and the representative image selection condition is “select an image with many common characters appearing”. For example, an image in which four or more common characters only appear is selected. The representative image selection condition 2 corresponds to the recommendation reason 2, and the representative image selection condition is "select according to the event type".

  In detail, the representative image selection condition is selected in accordance with the event type. The content is to select "an image showing the scenery if it is a trip", "an image showing a person if it is an everyday", or "an image showing a plurality of persons if it is a party" . The representative image selection condition 3 corresponds to the recommendation reason 3, and the representative image selection condition is to “select an image with an old photographing date and time”. In the second embodiment, image data of a plurality of images are sorted according to shooting date and time, and the image with the oldest shooting date and time is stored in the HDD 109 so that the image is the first. Therefore, the image of the image data at the beginning of the data is selected as the image with the old shooting date and time.

  FIG. 14 is a flowchart in which a process of generating a slide show for each display position of the screen and displaying a representative image of each slide show is added to the operation flow of recommending the slide show of FIG. In the following description, the description of steps for executing the same processing as the slide show in FIG. 9, such as S902, is omitted.

  In S1401, the display position on the screen of the representative image for each recommendation reason is set. This setting is optional. For example, although "up", "center" and "down" can be set on the screen, in the second embodiment, "left", "center" and "right" are set in order on the screen Do.

  The event attribute information management unit 202 acquires attribute information of an event image group to which the slide show being reproduced belongs (S901). Next, the recommendation reason determination unit 204 of the information processing device 3 refers to the item of the display position in the recommendation reason table 1201, and matches the display position set in S1401 with “left”, “center”, and “right”. Determine the recommendation reason as the recommendation reason to be displayed to the user.

  The display position may be set to another display position in accordance with the aspect ratio or size of the screen, and the recommendation reason table 1201 may correspond a plurality of display positions to one recommendation reason. For example, in the case where “upper left” and “lower left” are set as display positions, any of them may be displayed as the recommended reason of being similar to the character of ID1. In S902 to S906, an event image group and an image are selected as in the first embodiment.

  The representative image selection unit 1101 acquires a representative image selection condition corresponding to the ID of the determined recommendation reason from the representative image selection condition table 1301 (S1402). The representative image selection unit 1101 selects an image having the image ID selected in S906 from the image table 401 and having image attribute information matching the acquired representative image selection condition as a representative image (S1403). The number of images to be selected is the number of images displayed as a representative of the slide show, and a plurality of images may be selected.

  Thereafter, the information processing device 3 repeats S1401 to S303 and S901 to S908 until selection of all the display positions is completed (S1404). When selection for all display positions is completed, the recommendation unit 207 displays the representative image selected in S1403 at each display position, and displays a slide show recommended for the user (S1405). Then, when the user selects any representative image, the recommendation unit 207 reproduces the corresponding slide show.

  FIG. 15 is an explanatory diagram of a screen for displaying a representative image and recommending a slide show in the second embodiment. The screen 1001 is the same as that of FIG. 10, and the screen 1501 displays representative images of slide shows based on the corresponding recommended reasons at each of a plurality of display positions. In the second embodiment, the information on the reason for recommendation is displayed as text, but this display may be omitted.

After reproducing the last image of the slide show being reproduced on the screen 1001, the recommendation unit 207 switches to the screen 1501 and displays a representative image at each display position.
As described above, according to the second embodiment, the user can select a slide show of interest by displaying the representative image of each slide show generated for various recommendation reasons. Therefore, it is possible to browse a large number of images taken without getting bored. In addition, since the display position and the recommendation reason correspond to each other, it becomes easy to intuitively understand why each slide show is recommended.

Third Embodiment
In the first embodiment, there is one selection condition corresponding to one recommendation reason. However, the selection conditions suitable for one recommendation reason are not necessarily one. Therefore, in the third embodiment, an example in which a plurality of selection conditions are associated with one recommendation reason is shown. The differences from the first embodiment will be described below.

  FIG. 16 shows an event image group selection condition table 1601 in the third embodiment. The event image group selection condition table 1601 is obtained by modifying the event image group selection condition table of FIG. 5 so that a plurality of event image group selection conditions are associated with one recommendation reason.

  The recommended reason 1 will be described. Two event image group selection conditions corresponding to the event image group selection conditions 1 and 2 correspond to each other. The event image group selection condition 1 is to “select an event image group having a number of common characters greater than or equal to a designated number”. Further, the event image group selection condition 2 is that “select an event image group having a number of characters of a common child greater than a designated number”.

  The recommendation event reason 2 will be described. Two event image group selection conditions corresponding to the event image group selection conditions 3 and 4 correspond to each other. The event image group selection condition 3 is to “select an event image group having similar types of events”. In addition, the event image group selection condition 4 is that “a group of event images having a number of characters common to a child is greater than a designated number is selected”.

  To explain the recommendation reason 3, two event image group selection conditions corresponding to the event image group selection conditions 5 and 6 correspond to each other. The event image group selection condition 5 is "select an event image group whose difference in time of the shooting start date is within two weeks". The event image group selection condition 6 is that "an event image group having a two week difference in time of the shooting start date and a two hour time difference or less is selected".

  FIG. 17 shows an image selection condition table 1701 in the third embodiment. The image selection condition table 1701 is a modification of the event image group selection condition table 601 of FIG. 6 so that a plurality of image selection conditions are assigned to one recommendation reason.

  As illustrated, for the recommendation reason 1, two image selection conditions of image selection conditions 1 and 2 correspond. The image selection condition 1 is to “select an image showing a common character”. The image selection condition 2 is to “select an image in which two or more common characters appear together”.

  Two recommendation conditions, image selection conditions 1 and 2, correspond to recommendation reason 1. The image selection condition 1 is to “select an image showing a common character”. The image selection condition 2 is to “select an image in which two or more common characters appear together”.

  The two image selection conditions of image selection conditions 3 and 4 correspond to the recommended reason 2. The image selection condition 3 is "select the ratio of landscape and person according to the event type", and in detail, if the event is a trip, the image is 50% of the landscape image and 50% of the person image. Is selected. In the daily life, the image is selected so that the portrait image is 80% in the landscape image 20%, and the landscape image is 0% and the portrait image 100% in the party. The image selection condition 4 is the same as the image selection condition 3 in that "the ratio between the landscape and the person is selected according to the event type". However, in detail, if the event is a trip, the image is selected such that the landscape image is 100%, the person image is 100% in the daily life, and the person image is 100% in the party.

  Two recommended image selection conditions such as the image selection conditions 5 and 6 correspond to the recommended reason 3. The image selection condition 5 is "randomly select an image", and the image selection condition 6 is "select an image of a subject suitable for the season".

  FIG. 18 is a flowchart showing an operation of recommending a slide show using the tables shown in FIGS. 16 and 17. In S901 to S902, as in the first embodiment, the attribute information of the event image group to which the slide show being reproduced belongs is acquired, and the recommendation reason is determined.

  The event image group selection unit 205 acquires a plurality of event image group selection conditions corresponding to the recommendation reason determined in S902 from the event image group selection condition table 1601 shown in FIG. 16 (S1803). It is possible to arbitrarily determine which of the plurality of selection conditions is to be selected. For example, one of a plurality of selection conditions may be selected at random, or one may be selected in the order of small numbers or large numbers of event image group selection condition IDs.

  The event image group selection unit 205 selects an event image group for the acquired event image group selection condition (S1804). Thereafter, the image selection unit 206 acquires the image group condition based on the recommendation reason (S1805).

  In S906 and S907, as in the first embodiment, an event image group and an image are selected for each acquired image group condition, and a slide show is generated and recommended using the selected image. In the third embodiment, a plurality of event image group selection conditions and image selection conditions are associated with one recommendation reason. However, as in the second embodiment, the representative image selection condition managed by the rule management unit 208 may be acquired for each of the plurality of recommendation reasons.

  As described above, according to the third embodiment, by making a plurality of selection conditions correspond to one recommendation reason, various selections can be made in a range suitable for the recommendation reason, and the user can select a large number of images. You can browse without getting tired.

The first to third embodiments have been described in detail using the information processing apparatuses 1 and 3 as described above. However, the present invention can also be implemented, for example, in an embodiment as a system, an apparatus, a method, a program, or a storage medium.
Specifically, the present invention may be applied to a system constituted by a plurality of devices (for example, host computer, interface device, imaging device, web application, etc.), and may be applied to a device comprising one device .

  The object of the present invention is also achieved by the following. That is, a storage medium or storage medium storing a program code (computer program) of software for realizing the functions of the above-described embodiments is supplied to a system or apparatus. Such storage media are, of course, computer readable storage media. Then, the computer (or CPU or MPU) of the system or apparatus reads out and executes the program code stored in the storage medium. In this case, the program code itself read out from the storage medium realizes the functions of the above-described embodiments, and the storage medium storing the program code constitutes the present invention.

Claims (14)

Event image group recording means in which a plurality of event image groups each including a plurality of images associated with image attribute information are recorded;
A recommended reason recording means in which a recommended reason for recommended content recommended to be presented to the user after the content to be reproduced by the user is reproduced;
Event image group selection means for selecting an event image group to be used for generating the recommended content from the event image group using the recommendation reason;
From the images included in the selected event image group, the image attribute information is referenced to obtain the image satisfying the recommendation reason, and the image is used to generate the recommended content from the obtained images. And image selection means for selecting the selected image.
Information processing device. It further comprises a recommendation means for outputting to the presentation means the recommendation reason used for selecting the event image group.
An information processing apparatus according to claim 1. The recommendation means is characterized in that the recommendation content is generated from the image selected by the image selection means and output to the presentation means.
The information processing apparatus according to claim 2. Attribute information is associated with each of the content to be reproduced and the event image group,
The event image group selection means refers to the attribute information associated with the reproduced content and the attribute information associated with the event image group using the recommendation reason, and the reproduced content Select the event image group for
The information processing apparatus according to any one of claims 1 to 3. The apparatus further comprises condition recording means for recording an event image group selection condition which is a condition for selecting the event image group,
The event image group selection unit acquires, from the condition recording unit, the event image group selection condition corresponding to the recommendation reason used for generating the recommended content, and uses the acquired event image group selection condition. Select the event image group,
The information processing apparatus according to any one of claims 1 to 3. It further comprises condition recording means in which an image selection condition which is a condition for selecting the image is recorded,
The image selection unit acquires the image selection condition corresponding to the recommendation reason used for generating the recommended content from the condition recording unit, and selects the image using the acquired image selection condition ,
The information processing apparatus according to any one of claims 1 to 3. The image processing method further includes representative image selection means for selecting a representative image representative of the recommended content from the image selected by the image selection means based on the image attribute information and the recommendation reason.
The recommendation means further outputs the representative image to the presentation means.
The information processing apparatus according to claim 2. The recommendation means outputs the representative image to the presenting means so that the representative image is presented at a position on the screen defined for each of the recommendation reasons.
The information processing apparatus according to claim 7. A plurality of the recommended reasons are recorded in the recommended reason recording means,
The system further includes recommendation reason determination means for determining a plurality of recommendation reasons for selecting the event image group from a plurality of the recommendation reasons.
The image selecting means further selects the image for each of the determined plurality of recommendation reasons,
The representative image selecting means further selects the representative image for each of a plurality of the recommended contents generated for each of the determined plurality of recommendation reasons,
The recommendation means is configured to present both the determined recommendation reasons and the representative image of the recommended content generated by the recommendation reasons at a position on the screen defined for each of the determined recommendation reasons. And outputting the representative image to the presenting means.
The information processing apparatus according to claim 7. A plurality of the recommended reasons are recorded in the recommended reason recording means,
Furthermore, it is characterized by having a recommendation reason determination means for randomly determining a recommendation reason for selecting the event image group among a plurality of the recommendation reasons.
The information processing apparatus according to any one of claims 1 to 9. The recommendation reason recording means records a plurality of recommendation reasons,
Furthermore, it is characterized by having recommendation reason determination means which determines the recommendation reason for selecting the said event image group out of several said recommendation reasons based on the attribute information of the said content reproduced | regenerated.
The information processing apparatus according to any one of claims 1 to 9. In the event image group including a plurality of images associated with image attribute information using content recommendation reason recommended to be presented to the user after the content is reproduced for the content to be reproduced by the user Select the event image group to be used for generating the recommended content from
From the images included in the selected event image group, the image attribute information is referenced to obtain the image satisfying the recommendation reason, and the image is used to generate the recommended content from the obtained images. Selecting the image to be
Information processing method. Computer,
Event image group recording means in which a plurality of event image groups each including a plurality of images associated with image attribute information are recorded;
Recommendation reason recording means in which recommendation reason of recommendation content recommended to be presented to the user after the content to be replayed by the user is recorded,
Event image group selecting means for selecting an event image group to be used for generating the recommended content from the event image group using the recommendation reason;
Among the images included in the selected event image group, the image attribute information is referenced to obtain an image satisfying the recommendation reason, and the image used to generate the recommended content from among the determined images. Image selection means for selecting an image;
Computer program to function as.   A computer readable storage medium storing the computer program according to claim 13.