JP6534709B2 - Content information providing apparatus, content display apparatus, data structure of object metadata, data structure of event metadata, content information providing method, and content information providing program - Google Patents

Description

  The present invention provides a content information providing apparatus for providing information of a plurality of videos related to a predetermined program, a content display apparatus, a data structure of object metadata generated by the content information providing apparatus and a data structure of event metadata, and the content The present invention relates to a content information provision method and content information provision program corresponding to an information provision device.

  With the development of broadcasting technology, in relays for the purpose of a predetermined program such as a sporting event, a plurality of cameras may be installed, and programs may be photographed at various angles from each camera. In addition, the analysis information of the video and the information of the event in such a program as well as the environment in which it is released in parallel with the relay are prepared, and various video distribution is realized.

  In addition, with the use of information devices and communication technologies, users can acquire a wide variety of videos, and the options for videos to be viewed by the users are on the rise. In addition, a multi-screen environment becomes familiar due to a tablet or the like, and a user may view one event on a plurality of screens.

  With such technological progress, the user wants to select and view a desired event or object from a large number of images instead of the conventional broadcast type pre-edited and organized content. There is a growing demand for multi-content services.

  Currently, metadata in such video distribution is standardized by TVA (TV Anytime Forum), MPEG-7 (see Non-Patent Document 1), MPEG-21 and the like. These metadata manage, for each content, time-series information on the object such as the display position and display time of the object. Therefore, the user can extract a desired object from a plurality of videos by acquiring metadata on the content of all the related videos, and can simultaneously cope with multi-content viewing using a plurality of videos.

"INTERNATIONAL STANDARD Information technology-Multimedia Content Description interface Part 3 Visual", ISO / IEC 15983-3, May 15, 2002

  However, the conventional metadata standard has a problem that it is difficult to realize multi-content service. Specifically, since the user's interest extends to a plurality of contents, it is necessary to continuously and cross-search information such as objects sequentially updated in each of the plurality of contents, and the problem that real-time property is lost There is. In addition, even in the case of archive-type viewing, it is necessary to acquire information of the necessary content after acquiring all metadata related to a plurality of contents. Therefore, when the metadata of the content includes time-series information about the object such as the display position and the display time of the object, the resources used for the difference in acquisition of the content metadata become enormous, and the resource is wasted. is there.

  In addition, information on events needs to be embedded as text in metadata of each content, which makes it difficult to search for events and difficult to hold information on events not directly related to video.

  Therefore, an object of the present invention is to provide a content information providing apparatus, a content display apparatus, a data structure of object metadata generated by the content information providing apparatus, and data of event metadata which can easily select a user's desired video from a plurality of videos. It is an object of the present invention to provide a structure, a content information providing method and a content information providing program corresponding to the content information providing apparatus.

  In order to solve the above problems, a first feature of the present invention relates to a content information providing device for providing information of a plurality of videos related to a predetermined program. A content information providing apparatus according to a first feature of the present invention stores video analysis data in which an identifier of video data, detected object information of an object detected in the video data, and an object appearance time in the video data are associated. An object meta that assigns a storage device and an identifier of an object appearance that identifies a series of appearances of objects in video data, and associates an identifier of an object appearance with an identifier of video data, detected object information, and an object appearance time. Object meta generation means for generating object metadata including records.

  A second feature of the present invention relates to a content information providing apparatus for providing information of a plurality of videos related to a predetermined program. In the content information providing device according to the second aspect of the present invention, the storage device further includes an identifier of the video data, a shooting time of the video data, content-accompanying data corresponding to shooting object information, and an event generated in the program. A storage device that stores event information, event occurrence time, and status information data that associates related object information, and an identifier for the event, as well as for content-accompanying data, shooting time overlaps with event occurrence time And identifies the identifier of the video data to which the shooting object information and the related object information correspond, and the event occurrence time of the situation information data, the related object information corresponding to the shooting object information, and the identifier of the specified video data Event meta records that correspond to It comprises an event metadata generation means for generating an event-free metadata.

  The storage device further stores an identifier of the video data, detected object information of an object detected in the video data, and video analysis data in which object appearance time in the video data is associated, and a series of appearances of objects in the video data Object meta generation means for generating an object metadata including an identifier of an object appearance to be identified and an object appearance record including an identifier of video data, detected object information, and an object appearance time in which an object appearance time is associated with the identifier of the object appearance The event meta generation means is an object meta record common to the identifier of the video data of the event meta record and the related object information from the object metadata, To identify the object meta record the vent time of occurrence and the object appearance time overlaps, in the event meta record, the object appearance of the identifier of the specified object meta record may be associated with each other.

In the content-accompanying data, when there is a record corresponding to the identifier of the video data of the event meta record and the related object information, the identifier of the new object appearance is numbered and the identifier of the video data is added to the new object appearance identifier The information processing apparatus may further include additional object meta generation means for generating additional object metadata including related object information and an additional object meta record in which an event occurrence time of an event meta record is associated as an object appearance time.

  The storage device stores an identifier of the video data, detected object information of an object detected in the video data, and video analysis data in which object appearance time in the video data is associated, and identifies a series of appearances of objects in the video data Object meta generation means for generating object metadata including an identifier of an object appearance, an object appearance identifier, an identifier of video data, detected object information, and an object meta record in which an object appearance time is associated with the identifier of the object appearance; For the object meta record and the additional object meta record, the identifier of the video data and the object information are common, the object appearance time of the object meta record, and the object of the additional object meta record If transfected appearance time and overlap, the additional object metadata may further include an object metasynthesis means for deleting additional object meta records.

  The storage device further stores link master data that associates a search keyword with the identifier of the video data and the object information, and the event meta generation means stores the identifier of the video data of the event meta record and the object information in the event meta record. A search keyword corresponding to may be associated.

  According to a third aspect of the present invention, an object associated with a query input from a user is identified from object metadata acquired from the content information providing apparatus according to the first aspect of the present invention, and the identified object is identified. The identifier of the related video data and the object appearance time are specified, the identifier of the specified video data and the video data corresponding to the object appearance time are acquired, or from the content information providing device according to the second feature of the present invention From the acquired event metadata, identify the event related to the query input from the user, identify the identifier of the video data related to the identified event and the event occurrence time, identify the identified video data and the event occurrence Acquire video data corresponding to time, and display the acquired video data It is a content display device.

  A fourth aspect of the present invention relates to the data structure of object metadata associated with an object detected in video data. The data structure of the object metadata according to the fourth feature of the present invention is the identifier of the video data, the identifier of the object, and the video data, with the identifier of the object appearance as the key for identifying a series of the objects appearing in the video data The object meta record including the object appearance time specifying the time when the object is detected is included, and is used for the process of specifying the identifier of the video data related to the query input from the user and the object appearance time.

  A fifth aspect of the present invention relates to the data structure of event metadata associated with an event generated by a predetermined program for capturing video data. The data structure of the event metadata according to the fifth aspect of the present invention includes the event occurrence time, the object information of the event, and the occurrence time of the event in the shooting time, using the event identifier as a key, and the event object It includes an event meta record that associates an identifier of video data included as a shooting target, and is used for processing of identifying an identifier of video data related to a query input from a user and an event occurrence time.

  A sixth aspect of the present invention relates to a content information providing method for providing information of a plurality of videos related to a predetermined program. In a content information providing method according to a sixth aspect of the present invention, a computer associates video analysis data in which an identifier of video data, detected object information of an object detected in the video data, and an object appearance time in the video data are associated. , Storing, and an identifier of an object appearance that identifies a series of appearances of the object in the video data, and an identifier of the object appearance, an identifier of the video data, detected object information, and an object appearance time Generating object metadata including the associated object meta record.

  A seventh aspect of the present invention relates to a content information providing method for providing information of a plurality of videos related to a predetermined program. In a content information providing method according to a seventh aspect of the present invention, a computer adds an identifier of video data, a shooting time of video data, content-accompanying data in which shooting object information is associated, and an event name of an event generated in a program , Storing event information, status information data associated with related object information, and the computer assigns an identifier of the previous event, and in the content-accompanying data, the shooting time overlaps with the event occurrence time And identifies the identifier of the video data to which the shooting object information and the related object information correspond, and the event occurrence time of the situation information data, the related object information corresponding to the shooting object information, and the identifier of the specified video data Event event that Comprising the step of generating an event metadata including records.

An eighth aspect of the present invention is a content information provision program for causing a computer to function as the content information provision device according to the first or second aspect of the present invention.
A ninth aspect of the present invention relates to a content information providing system for providing information of a plurality of videos related to a predetermined program. An information providing system according to a ninth aspect of the present invention stores an identifier of video data, detected object information of an object detected in the video data, and video analysis data in which object appearance time in the video data is associated. An object meta record in which an identifier of an object appearance that identifies a device and a series of appearances of objects in video data is numbered, and an identifier of an object appearance is associated with an identifier of video data, detected object information, and an object appearance time. Content information providing apparatus including object meta generation means for generating object metadata including the content information display apparatus and content information display apparatus for displaying video data specified by an object meta record related to a predetermined query Provided with a location.
A tenth aspect of the present invention relates to a content information providing system for providing information of a plurality of videos related to a predetermined program. According to a ninth aspect of the present invention, there is provided an information providing system comprising: an identifier of video data, a shooting time of video data, content-accompanying data correlated with shooting object information, an event name of an event generated by the program, an event occurrence time , And a storage device for storing status information data associated with related object information, and an identifier of an event numbered, and in the content accompanying data, the shooting time overlaps with the event occurrence time, and the shooting object information is related An event meta record in which the identifier of the video data to which the object information corresponds is specified, and the identifier of the event is associated with the event occurrence time of the situation information data, the related object information corresponding to the shooting object information, and the identifier of the specified video data Raw event metadata including Comprising a content information providing apparatus comprises an event metadata generation means for, in the content information providing device, identified by the event metadata records associated with a given query, the content display device for displaying video data.

  According to the present invention, a content information providing apparatus capable of easily selecting a video desired by the user from a plurality of videos, a data structure of object metadata generated by the content information providing apparatus and a data structure of event metadata, and the content information A content information provision method and a content information provision program corresponding to the provision device can be provided.

It is a conceptual diagram explaining the data organization of metadata concerning an embodiment of the invention. It is a conceptual diagram explaining the data organization of general metadata. It is a flowchart explaining the outline | summary of a process of the content information provision method which concerns on embodiment of this invention. BRIEF DESCRIPTION OF THE DRAWINGS It is a figure explaining the system configuration | structure of the content information provision system which concerns on embodiment of this invention, the hardware constitutions of a content information provision apparatus, and a functional block. It is a figure explaining an example of a data structure of image analysis data concerning an embodiment of the invention. It is a figure explaining an example of a data structure of content incidental data concerning an embodiment of the invention. It is a figure explaining an example of a data structure of situation information data concerning an embodiment of the invention. It is a figure explaining an example of a data structure of link master data concerning an embodiment of the invention. It is a figure explaining an example of a data structure of object metadata concerning an embodiment of the invention. It is a figure explaining an example of a data structure of event metadata concerning an embodiment of the invention. It is a flow chart explaining object meta generation processing concerning an embodiment of the invention. It is a figure explaining an example of data in object meta generation processing concerning an embodiment of the invention. It is a flow chart explaining event meta generation processing concerning an embodiment of the invention. It is a figure explaining an example of the data in the event meta production | generation process which concerns on embodiment of this invention. It is a flowchart explaining the additional object meta generation process which concerns on embodiment of this invention. It is a figure explaining an example of data in additional object meta generation processing concerning an embodiment of the invention. It is a flow chart explaining object meta integration processing concerning an embodiment of the invention. It is a figure explaining an example of data in object meta integrated processing concerning an embodiment of the invention. It is a flowchart explaining the event meta replenishment process which concerns on embodiment of this invention. It is a figure explaining an example of the data in the event meta replenishment process which concerns on embodiment of this invention. It is a flow chart explaining offer processing concerning an embodiment of the invention.

Next, embodiments of the present invention will be described with reference to the drawings. In the description of the following figures,
The same or similar parts are given the same or similar symbols.

(Data system of metadata)
With reference to FIG. 1, a data system of metadata of content according to the embodiment of the present invention will be described. The metadata shown in FIG. 1 is provided for a plurality of video contents related to a predetermined program such as, for example, relay of a sport event, relay of a concert, and the like.

  The metadata includes metadata for each content, object metadata 21 and event metadata 22. In the embodiment of the present invention, each metadata is described as XML, but it is not limited thereto.

  In the embodiment of the present invention, an "object" is a substance having a size and a shape, and is an object which can identify a position existing in a space. For example, specific "objects" in a baseball game are places where baseball players such as pitchers and batter, and baseball players such as batter boxes, benches, and bullpens pitch. The "event" is a trigger or an event that changes the state, and the appearance, the condition, and the state are different before and after the "event". For example, a specific “event” in a baseball game is a series of matches that are decided by the team, the state of offense and defense changes such as back and forth once, the state in which the batter enters the batter box, the batter hits It is the timing that I hit.

  As metadata for each content, a CRID (Content Reference ID) for identifying the content (video data) is associated with data that is not influenced by time series such as an outline of the content, a cast, and the like. Metadata for each content is generated based on content-accompanying data 12 described later. Time-series information such as the display position of an object, display time, and information on events is not included in the metadata for each content, but is included in the object metadata 21 and event metadata 22.

  The object metadata 21 is an object appearance identifying a time when an object is detected in video data, an object ID (object information), and a video data, using an object appearance ID identifying a series of appearances of an object in video data as a key. It has a data structure that gathers object meta records that associate time and occurrence position. The object metadata 21 is generated based on video analysis data 11 obtained by technically analyzing a video. The object occurrence ID identifies a series of occurrences of the object in the video data. Here, the appearance of a series of objects is a partial time interval in which an object is displayed in a video due to a break on a video representation such as a scene or a cut, movement of a camera angle of view, or missing due to movement of an object.

  By specifying the CRID of the video data in which the object appears in the object metadata 21, the metadata for each content and the object metadata 21 can be mutually referred to, and can mutually complement information. . Further, by searching the object metadata 21 using the object ID as a query, it becomes possible to complement information on the appearance of a predetermined object in the object metadata 21.

  The event metadata 22 is an event meta record that associates an event occurrence time, object information of the event, and a CRID including the object of the event as a shooting target with the event ID as a key and including the event occurrence time as the shooting target. It has a data structure to aggregate. The event metadata 22 is generated based on the situation information data 13 provided by a program provider or the like. The event ID identifies an event that has occurred in a program to be shot.

  By specifying the CRID of the video data related to the event in the event metadata 22, the metadata for each content and the event metadata 22 can be mutually referred to, and can mutually complement information. . In addition, information on a predetermined object or time can be complemented in the event metadata 22 by searching event metadata using an object related to an event, an occurrence time, and the like as a query.

  Here, in the object metadata 21, a new object meta record is generated each time appearance of a series of objects is specified, and in the event metadata 22, a new event meta record is generated each time an event occurs. Ru. Therefore, the user can select and view desired content without impairing real-time performance. In addition, the user can view and listen to a video related to a desired object or event without referring to a large amount of data.

  On the other hand, in general metadata of content shown in FIG. 2, object information and event information are associated with CRID as a key. This metadata stores all the information about the content. Therefore, such general content metadata has a large amount of information, may impair the real-time capability, and may press resources.

(How to provide content information)
A content information provision method according to the embodiment of the present invention will be described with reference to FIG. In FIG. 3, in the content information providing system 5 shown in FIG. 4, the case where the content information providing device 1 generates the object metadata 21 and the event metadata 22 and provides the generated data to the content display device 3 is assumed. explain.

  In step S1, the content information providing device 1 generates an object meta record from the video analysis data 11 and updates the object metadata 21 as the program progresses. The process of step S1 is executed each time a series of object occurrences occur.

  In step S2, the content information providing device 1 generates an event meta record from the status information data 13 and updates the event metadata 22 as the program progresses. The process of step S2 is executed each time the status information data 13 is updated.

  In step S3, the content information providing device 1 generates an additional object meta record, paying attention to the object information of the event meta record.

  In step S4, the content information providing apparatus 1 supplements the event meta record generated in step S2 with the information of the event meta record of the event having a parent-child relationship with this event.

  The additional object meta record generated in step S3 may overlap with the object meta record generated in step S1. Therefore, in step S5, the information of the additional object meta-record is integrated into the object meta-record as necessary, and the integrated additional object meta-record is deleted.

  When the object meta record and the event meta record are generated in steps S1 to S5, the content information providing device 1 provides the generated record to the content display device 3 in step S6. For example, when the content display device 3 inputs a predetermined query, a record related to the query is provided to the content display device 3.

(Content information provision system)
The content information providing system 5 according to the embodiment of the present invention will be described with reference to FIG. The content information providing system 5 includes a content information providing device 1, a content distribution device 2 and a content display device 3. These devices are communicably connected to each other via a communication network 4 such as the Internet.

  The content information providing device 1 provides information of a plurality of videos related to a predetermined program. The content information providing device 1 generates metadata about the content and provides the content display device 3 with the metadata.

  The content distribution device 2 provides a plurality of video data related to a predetermined program. The content distribution device 2 provides video data to the content display device 3 in accordance with the designated CRID, time, and the like. The content distribution device 2 provides video data regarding each CRID and time even when a plurality of CRIDs and time are specified.

  The content display device 3 acquires each record of the object metadata 21 and the event metadata 22 from the content information providing device 1. The content display device 3 identifies a desired object or event based on an object name or event name input from the user or a search keyword, and the user desires to view the content based on the object or event desired by the user. Identify the CRID and time. The specified information is input to the content distribution device 2, and desired content is acquired from the content distribution device 2.

  The configuration shown in FIG. 4 is an example, and the present invention is not limited to this. For example, the content information providing device 1 and the content distribution device 2 may be implemented in the same device. In addition, the content information providing device 1 acquires information of an object or event desired by the user from the content display device 3, specifies a CRID or time that the user desires to view, and provides it to the content display device 3. Also good.

(Content information provision device)
The content information providing device 1 is a general computer including storage devices 10 and 20, a processing device 30, a communication control device (not shown), and the like. Each function shown in FIG. 1 is realized by executing a content information providing program for realizing a predetermined function by a general computer.

  In FIG. 4, the storage device 10 stores data for storing data input to the content information providing device 1, and the storage device 20 stores data output from the content information providing device 1. It is not limited to such a configuration.

  The storage device 10 stores video analysis data 11, content-accompanying data 12, status information data 13, and link master data 14.

  The video analysis data 11 is data obtained by analyzing the time when a specific object appears in the content distributed by the content distribution device 2. As shown in FIG. 5, the video analysis data 11 associates a CRID (identifier of video data), detected object information of an object detected in the video data, and object appearance information in the video data. Here, the detected object information is information of an object detected from video data. In the embodiment of the present invention, the object information is information including the name of the object and the object ID, and is linked with the parent-child relationship, the attribute, etc. of the object by object dictionary data (not shown). The object appearance information is information including a time and an area in which an object is detected in predetermined video data by video analysis.

  The content-accompanying data 12 is data relating to an attribute of video data. As shown in FIG. 6, the content-accompanying data 12 associates a CRID (video data identifier), video data shooting time, and shooting object information. The content-accompanying data 12 may further be associated with information such as the shooting purpose of the video data and the installation place of the camera.

  The situation information data 13 is data input from a system such as a baseball player or the like who performs content programming, and is information of an event that has occurred in the program. As shown in FIG. 7, the status information data 13 associates an event name of an event generated in a program, an event occurrence time, and related object information. Here, related object information is information such as an object name and an object ID of an object related to an event. The status information data 13 may further correspond to a parent event ID for this event.

  An event may be progressed at the same time, and a part of the time interval of one event may be defined as another event. If it is considered that this part of the event is a component of the event that includes the time interval, the parent-child relationship of the event is established. Specifically, a parent-child relationship is established between the “3rd Table” event and the “3rd Table First At bat” event in a baseball game. Therefore, in the status information data 13, the ID of the parent event for the generated event is set.

  The link master data 14 is data generated in advance by a program player or the like, and has a function of filtering or adding a search keyword so that an object or event desired by the user can be easily searched. As shown in FIG. 8, the link master data 14 associates a CRID (image data identifier), an event name, and linking object information related to the event name. The link master data 14 may also associate a search keyword with the CRID and the linking object information.

  The storage device 20 stores object metadata 21, event metadata 22 and additional object metadata 23.

  The object metadata 21 is data generated by the object meta generation unit 31 with respect to a series of appearances of a predetermined object in predetermined video data. As shown in FIG. 9, the object metadata 21 is data including a plurality of object meta records in which object information, CRID, object appearance information and parent object ID are associated with an object appearance ID as a key.

  The event metadata 22 is data generated by the event meta generation unit 32 for an event generated in a program. As shown in FIG. 10, the event metadata 22 is a plurality of event meta data in which CRID, event occurrence time, event name, related object information, object appearance ID, parent event ID, and search keyword are associated with event ID as a key. It is data that contains a record.

  The additional object metadata 23 is data having the same data configuration as the object metadata 21, and is generated by the additional object meta generation unit 33.

  The processing device 30 includes an object meta generation unit 31, an event meta generation unit 32, an additional object meta generation unit 33, an object meta integration unit 34, an event meta replenishment unit 35, and a provision unit 36.

(Object meta generation means)
The object meta generation unit 31 generates object metadata 21.

  The object meta generation unit 31 refers to the video analysis data 11 to assign an object appearance ID for identifying a series of appearances of the object in the video data, and, in the object appearance ID, CRID, object information, and object appearance time To create an object meta record. The object meta generation unit 31 generates object metadata including the generated object meta record. Here, the object meta generation unit 31 may associate object related information such as a parent object ID with reference to object dictionary data.

  The processing by the object meta generation unit 31 according to the embodiment of the present invention will be described with reference to FIGS. 11 and 12. As shown in FIG. 12, the video analysis data 11 indicates that the object “OBJID_1” appears in a series of time “OAT_1” of the CRID “CRID_1”.

  First, in step S101, the object meta generation unit 31 refers to the video analysis data 11, and assigns an object appearance ID "OA_1" for a series of object appearances. In step S102, the object meta generation unit 31 adds the CRID “CRID_1” of the video analysis data 11, the object information “(pitcher, OBJID_1)”, and the object appearance information “(OAT_1” to the object appearance ID “OA_1” numbered in step S101. , OAR_1) ”. In step S102, the parent object ID "POBJID_1" is specified from the object information, and is associated with the CRID "CRID_1".

  In step S104, the object meta record generated in step S102 and step S103 in association with the object appearance ID is inserted into the object metadata 21.

(Event meta generation means)
The event metadata generation means 32 generates event metadata 22.

  The event meta generation unit 32 refers to the status information data 13 to number the event ID, and in the content attached data 12, the shooting time includes the event occurrence time, and the shooting object information corresponds to the related object information. Identify the CRID. The event meta generation unit 32 generates an event meta record in which the event occurrence time of the status information data 13, the related object information corresponding to the shooting object information, and the specified CRID are associated with the event ID.

  Here, the event meta generation unit 32 may generate an event meta record when the object information corresponding to the specified CRID and the event name in the link master data 14 is associated with the imaging object information. For example, in the link master data 14, the information of the object related to the event is set for the event that the performer or the like wants to show to the user. As a result, even when an object closely related to an event and a related shallow object are set in the status information data 13, the link master data generates an event meta record on the related related deep object and the related shallow Control can be made to not generate event meta records for objects.

  Furthermore, the event meta generation unit 32 assigns an object appearance ID to the event meta record. Specifically, the event meta generation unit 32 is an object meta record common to the CRID of the event meta record and the related object information from the object metadata 21 and is an object in which the event occurrence time and the object appearance time of the event meta record overlap. Identify meta records. The event meta generation unit 32 associates the event occurrence ID of the specified object meta record with the event meta record. The event meta generation unit 32 associates an object appearance ID with an object appearance ID, paying attention to the appearance of the event object in the video data when the event occurs. The event meta generation means 32 specifies the CRID, the time of occurrence of the event, and the object meta record in which the objects overlap from the object metadata 21 and associates the object appearance ID of the specified record with the event meta record.

  Furthermore, the event meta generation unit 32 may associate the search keyword with the event meta record. The event meta generation unit 32 refers to the link master data 14 to extract a record corresponding to the CRID and event name of the event meta record, and associates the search keyword associated with the object information with the event meta record.

  The event meta generation unit 32 may associate CRIDs of a plurality of video data with the event ID, or may correspond an identifier of a program for bundling video data.

  The event meta generation unit 32 generates the event metadata 22 including the event meta record generated as described above.

  The processing by the event metadata generation unit 32 according to the embodiment of the present invention will be described with reference to FIGS. 13 and 14. As shown in FIG. 14, in the situation information data 13, an event in which the batter hits is described.

  First, in step S201, the event meta generation unit 32 refers to the content-accompanying data 12 for the event of the situation information data 13, and the shooting time of the content-accompanying data 12 overlaps with the event occurrence time "IT_1" of the situation information data 13. The shooting object information of the content-accompanying data 12 searches for a record corresponding to the related object information “(pitcher, OJID_1) (butter, OBJID_2)” of the situation information data 13. Here, the event meta generation unit 32 specifies, from the content-accompanying data 12, the CRID “CRID_1” associated with the imaging object information “(pitcher, OJID_1)” at the imaging time “CT_1”. Here, the imaging time “CT_1” overlaps at least a part of the event occurrence time “IT_1”.

  In step S202, the event meta generation unit 32 acquires, from the link master data 14, the linking object information “(pitcher, OJID_1) (butter, OBJID_2)” corresponding to the CRID “CRID_1” and the event name “hit”.

  In step S 203, the event meta generation unit 32 determines whether or not the associating object information of the link master data 14 corresponds to the photographing object information of the content accompanying data 12. When the linking object information of the link master data 14 does not correspond to the shooting object information of the content-accompanying data 12, the video data of this CRID “CRID_1” captures the time at which the hit event occurred, but It is determined that the object that the user wants to show is photographed, and the process ends.

  Here, in the case of the example shown in FIG. 14, the associating object information “(pitcher, OJID_1) (butter, OBJID_2)” of the link master data 14 and the imaging object information “(pitcher, OJID_1)” of the content accompanying data 12 The object information “(pitcher, OJID_1)” overlaps. Therefore, it is determined that the video data of this CRID “CRID_1” is photographing an object that the performer wants to show in the hit event.

  In this case, in step S204, the event ID is assigned, and the process of generating an event meta record is executed. Specifically, in step S205, the event ID is CRID "CRID_1" specified in step S201, event occurrence time "IT_1" of status information data 13, event name "hit", related object information "(pitcher, OJID_1 ) (Batter, OBJID_2) ”and the parent event ID“ PID_1 ”are associated.

  Furthermore, in steps S206 and S207, the event meta generation unit 32 assigns an object appearance ID. The event meta generation means 32 includes object information including one of CRID “CRID_1” and object information “(pitcher, OJID_1) (butter, OBJID_2)” from the object metadata 21, and the event occurrence time and the object appearance time overlap. Identify the record to be acquired, and obtain the object appearance ID of that record. Here, when there is object metadata shown in FIG. 12B, the object appearance ID “OA_1” is associated with the event ID.

  In step S208, the event meta record generated in step S205 and step S207 in association with the event ID is inserted into the event metadata 22.

(Additional object meta generation means)
The additional object meta generation unit 33 generates additional object metadata 23. The additional object metadata 23 identifies the appearance information of the object related to the event from the information on the event, and generates an additional object meta record. The additional object meta-record is generated from the information on the event, while the object meta-record is different from that generated from the video analysis data 11.

  The additional object meta generation unit 33 assigns a new object appearance ID when there is a record corresponding to the CRID of the event meta record and the related object information in the content additional data 12. The additional object meta generation unit 33 generates an additional object meta record in which the CRID and related object information are associated with the new object appearance ID, and the event occurrence time of the event meta record is associated as the object appearance time. The additional object meta generation unit 33 generates the additional object metadata 23 including the additional object meta record generated as described above.

  The processing by the additional object meta generation unit 33 according to the embodiment of the present invention will be described with reference to FIGS. 15 and 16. As shown in FIG. 16, the case of generating an additional object meta record from the record corresponding to the event name “hit” and the related object “(pitcher, OJID_1) (butter, OBJID_2)” of the event metadata 22 will be described. .

  First, in step S301, it is determined whether the content-accompanying data 12 has a record corresponding to the CRID of the event meta-record and the related object information. In the example shown in FIG. 16, the event metadata 22 has a record of CRID “CRID_1” and related object information “(pitcher, OJID_1) (butter, OBJID_2)”, and the content attached data 12 has CRID “CRID_1” and photographing Since there is a record of object information "(pitcher, OJID_1)", both correspond. Therefore, in the case of the example shown in FIG. 16, it progresses to step S302. On the other hand, when there is no corresponding record, the processing is ended as it is.

  In step S302, the additional object meta generation unit 33 assigns an object appearance ID. In step S303, the additional object meta generation unit 33 associates the object appearance ID with CRID “CRID_1”, related object information “CRID“ CRID_1 ”, and related object information“ (pitcher, OJID_1) ”, and sets the event as an object appearance time. Correspond the occurrence time “IT_1”. Further, in step S304, a parent object ID is associated with the object appearance ID. The process of step S304 is the same as the process of step S103 of FIG.

  In step S305, the additional object meta record generated in step S303 and step S304 in association with the object appearance ID is inserted into the additional object metadata 23.

Here, when a plurality of related objects are associated with an event meta record,
Additional object meta records may be generated for each related object. In the example shown in FIG. 16, the object appearance ID may be assigned to the related object information “(butter, OBJID_2)” to generate additional object metadata. In addition, also in the generation of the additional object meta record, the link master data 14 or the like may be filtered so that the meta record expected by the sponsor is generated.

(Object meta integration means)
The object meta integration means 34 eliminates duplication of the object metadata 21 and the additional object metadata. The object metadata 21 is generated from the video analysis data 11, and the additional object metadata 23 is generated from the situation information data 13. Therefore, since each record may be generated redundantly, the object meta integration means 34 integrates each data.

  The object meta integration means 34 adds the CRID and the object information to the object meta record and the additional object meta record, and when the object appearance time of the object meta record and the object appearance time of the additional object meta record overlap, the additional object The additional object meta-record is deleted from the metadata 23.

  The object meta integration means 34 is executed after the processing by the event meta replenishment means 35, as shown in FIG.

  The processing by the object meta integration means 34 according to the embodiment of the present invention will be described with reference to FIGS. 17 and 18.

  First, in step S401, the object meta integration unit 34 refers to the object metadata 21 and the additional object metadata 23, and identifies records in which the CRID and the object information are common and the object appearance times overlap. The object meta record and the additional object meta record shown in FIGS. 18A and 18B have the same CRID “CRID_1”, and have the same object information “(pitcher, OJID_1)”. If the object appearance time “Ta1−Ta2” of the object meta record overlaps with the object appearance time “Tb1−Tb2” of the additional object meta record, the process proceeds to step S402.

  In step S402, the object appearance time Ta1 to Ta2 of the object meta record before integration and the object appearance time T1 to T2 of the object meta record after integration are changed as necessary. For example, when the object appearance time is determined based on the and condition of the object appearance time "Ta1-Ta2" of the object meta record and the object appearance time "Tb1-Tb2" of the additional object meta record, T1 is Ta1 and Tb1. T2 is the later of Ta2 and Tb2. On the other hand, when the object appearance time is determined based on the or condition, T1 is the earlier of Ta1 and Tb1, and T2 is the later of Ta2 and Tb2.

In step S403, the additional object meta record is deleted from the additional object metadata 23.
This eliminates duplication of object meta records and additional object meta records. Further, records of the additional object metadata 23 which do not overlap with the records of the object metadata 21 may be inserted into the object metadata 21.

(Event meta replenishment means)
The event meta replenishment means 35 replenishes the information of the event meta record based on the parent-child relationship of the event. When there is a child event meta record in which a parent event ID is set and a parent event meta record corresponding to the parent event ID, the event meta replenishment means 35 replenishes information on related objects and information on an occurrence time zone.

  The process by the event meta replenishment means according to the embodiment of the present invention will be described with reference to FIGS. 19 and 20. As shown in FIG. 20, since the event ID of the event meta record shown in FIG. 20 (a) is set in the parent event ID of the event meta record shown in FIG. 20 (b), it is shown in FIG. The event meta record event is a parent event of the event meta record event shown in FIG.

  First, in step S501, the event meta replenishing unit 35 extracts, from the event metadata 22, a record in which a parent event ID is set and a record of the parent event ID. Here, event meta records shown in FIGS. 20A and 20B are extracted.

  In step S502, object information is changed as necessary. For example, since the related object information in FIG. 20 (b) is null, as shown in FIG. 20 (c), the related object information in the new event meta record is the parent's event meta shown in FIG. 20 (a). Related object information of the record is set. This process is suitable when the relationship between related objects does not change, such as the relationship between a batting event and a hit event during batting.

  In step S503, the event occurrence time is changed as necessary. For example, when the event occurrence time of the child event metadata is not set, the event occurrence time of the parent event metadata may be set.

  As described above, the event meta supplementing means 35 supplements information based on the parent-child relationship of event meta records, and updates the event metadata 22.

(Provided means)
The providing means 36 outputs data of one of the object meta record and the event meta record. The providing unit 36 extracts a record suitable for the query from the object metadata 21 and the event metadata 22 based on the query input from the content display device 3 or the like, and outputs the extracted record.

  The content display device 3 acquires video data specified by the received object meta record or event meta record from the content distribution device 2 and displays it on a display device (not shown) or the like.

  The process by the providing means 36 will be described with reference to FIG.

  First, in step S601, the providing unit 36 acquires a query, and in step S602, extracts an object meta record and an event meta record corresponding to the input query. In step S603, the providing unit 36 outputs the extracted record.

  Alternatively, the providing means 36 may provide the object metadata 21 and the event metadata 22 to the content display device 3. The content display device 3 extracts a desired record from the acquired data, specifies desired video data, acquires it from the content distribution device 2, and displays it on a display device or the like.

(Action effect)
As described above, the content information providing apparatus 1 according to the embodiment of the present invention associates the metadata that is not influenced by the time series with the metadata for each content, and the object metadata 21 and the event meta for the data that changes in time series. Data 22 is generated. This allows the user to easily search for desired content on an object or event basis. Also, by adding a CRID to the object metadata 21 and the event metadata 22, it is possible to easily refer to video data from an object or event.

  In archive viewing, it is possible to easily create a playlist in which desired objects and events are collected.

  Even while the program is in progress (live), the object meta record and the event meta record are sequentially created along with the progress, so the user can view the desired video data without losing the live immediacy . In addition, even when new video data is added while the program is in progress (live), it is possible to flexibly cope with it without largely changing the processing of the system.

  In addition to generating an object meta record from the video analysis data 11, an additional object meta record is also generated from the status information data 13 (event metadata 22), so various object meta records can be generated. . Also, since the records of the predetermined condition are integrated by the object meta integration means 34, the information can be enriched without significantly increasing the amount of information of the object meta records.

  Furthermore, by defining an event parent-child relationship in the event meta record, information on child events is also provided when a parent event is searched, and information on parent events is provided when a child event is searched. It is also possible to enrich the search results. The event meta supplementing means 35 can also enrich the information of each record based on the parent-child relationship of the event.

  As described above, the content information providing device 1 according to the embodiment of the present invention makes it possible to easily select and reproduce a video desired by the user from a plurality of videos, both at the time of live and at the time of archive reproduction.

(Other embodiments)
As mentioned above, although described by the embodiment of the present invention, it should not be understood that the statement and drawing which make a part of this disclosure limit this invention. Various alternative embodiments, examples and operation techniques will be apparent to those skilled in the art from this disclosure.

  For example, the content information providing apparatus described in the embodiment of the present invention may be configured on one hardware as shown in FIG. 1, or on a plurality of hardware depending on the function and the number of processing. It may be configured. Also, it may be realized on an existing content delivery system.

  In the embodiment of the present invention, the content information providing apparatus generates and outputs both object metadata and event metadata. However, only one of them may be generated and output.

  Of course, the present invention includes various embodiments and the like which are not described herein. Accordingly, the technical scope of the present invention is defined only by the invention-specifying matters according to the scope of claims appropriate from the above description.

1 Content Information Providing Device 2 Content Distribution Device 3 Content Display Device 4 Communication Network 5 Content Information Providing System 10 Storage Device 11 Image Analysis Data 12 Content Associated Data 13 Status Information Data 14 Link Master Data 21 Object Metadata 22 Event Metadata 23 Addition Object metadata 31 Object meta generation means 32 Event meta generation means 33 Additional object meta generation means 34 Object meta integration means 35 Event meta replenishment means 36 Provision means

Claims (5)

A metadata structure including object metadata and metadata for each content, wherein the object metadata and the metadata for each content can be mutually referenced.
The object metadata is configured to include a plurality of object meta records.
The object meta records, the identifier of the video data which is information for identifying the video data, and object identifier which is information for identifying the object that is detected in the video data, the time in which the object appears in have you to the video data is constituted by containing an object appearing time is information specifying a
The metadata for each content includes an identifier of the video data and information on the video data which is not influenced by the time series of the video data.
The content information providing apparatus, the content distribution apparatus, or the content display apparatus specifies an object identifier corresponding to an object related to the query input from the user, specifies an object meta record including the object identifier, and includes it in the object meta record Rume metadata structure used in processing to identify a set of identifiers and object appearance time of the video data, identifies the metadata for each content containing an identifier of the video data.
A metadata structure including event metadata and metadata for each content, wherein the event metadata and the metadata for each content can be mutually referenced,
The event metadata is configured to include a plurality of event meta records.
The event metadata records, the identifier of the video data which is information for identifying the image data, and the event identifier is information for identifying the detected event in the video data, the time the event had contact to the video data is generated and event occurrence time information for specifying the configured includes,
The metadata for each content includes an identifier of the video data and information on the video data which is not influenced by the time series of the video data.
The content information providing apparatus, the content distribution apparatus, or the content display apparatus identifies an event identifier corresponding to an event related to a query input from the user, identifies an event meta record including the event identifier, and includes the event meta record Rume metadata structure used in processing to identify a set of identifiers and event occurrence time of the video data, identifies the metadata for each content containing an identifier of the video data.
The object meta record is a parent object identifier that is information specifying a parent object having a parent-child relationship with the object specified by the object identifier.
The metadata structure according to claim 1, further comprising:
The event meta record is a parent event identifier that is information specifying a parent event that includes the event specified by the event identifier as a component
The metadata structure according to claim 2, further comprising:
The event meta record is a related object identifier that is information specifying an object related to the event specified by the event identifier.
The metadata structure according to claim 4, further comprising:

Images